EP2202647A1 - Method and apparatus for organizing segments of media assets and determining relevance of segments to a query - Google Patents
Method and apparatus for organizing segments of media assets and determining relevance of segments to a query Download PDFInfo
- Publication number
- EP2202647A1 EP2202647A1 EP09180762A EP09180762A EP2202647A1 EP 2202647 A1 EP2202647 A1 EP 2202647A1 EP 09180762 A EP09180762 A EP 09180762A EP 09180762 A EP09180762 A EP 09180762A EP 2202647 A1 EP2202647 A1 EP 2202647A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- subject matter
- media
- ontology
- media item
- segment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/44—Browsing; Visualisation therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24578—Query processing with adaptation to user needs using ranking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/438—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
Definitions
- the invention pertains to a process and apparatus for organizing segments of audio, video, and media files and determining the relevance of such segments to each other or to a query.
- the only way a consumer of media content can access a segment of particular interest to that consumer within a media asset is to scan through the asset in a linear fashion, such as by using a fast-forward or rewind function of a media player, to find the desired content.
- Media refers to the forms in which content may be transmitted.
- audio e.g., music, speech
- visual photographs, drawings, motion pictures, web pages, animation
- These media are typically represented in electronic formats, such as, for example, HTTP, NNTP, UDP, JMS, TCP, MPEG, MP3, wave files, HTML, JPEG, TIFF, and PDF.
- transmitted media will likely involve other sensory data such as taste, smell and touch.
- Contextual information or “subject matter” refers broadly to the topic or theme of the content and can be virtually anything within the realm of human knowledge, such as baseball, strike out, fast ball, stolen base, mountains, scary, happy, George Carlin, nighttime, cool, winner.
- the nature and duration of each segment will depend, of course, on the particular ontology.
- advertisers often purchase advertisement time or space within media assets such as television programs, web pages, podcasts, and radio programs based on the subject matter of the media.
- media assets such as television programs, web pages, podcasts, and radio programs based on the subject matter of the media.
- advertisers commonly are interested in a particular demographic of media consumers that can range from the very broad to the extremely narrow. For instance, a producer of beer might be interested in a demographic of male media consumers aged 18-45, whereas a producer of anti-aging face cream for women might be interested in a demographic comprising female viewers aged 30-70.
- the subject matter of a media asset often has a very high correlation to a specific demographic.
- the producer of anti-aging face cream may be much more interested in placing its advertisement in the middle of a soap opera rather than a football competition because the soap opera will be viewed by many more individuals within the demographic that is likely to buy its product than the football competition, even if the football competition has a much larger overall viewing audience than the soap opera.
- media content providers e.g., individual television and radio stations, cable, fiber optic and satellite subscription-based television network operators, internet service providers
- media content producers e.g., television and radio program producers, podcasters, website operators
- advertisers in determining what subject matters of such media appeal to particular demographics for advertisement placement and other purposes.
- the invention pertains to methods, systems, and apparatus for identifying media items relevant to a subject matter of a first media item, the method comprising determining the subject matter of a first media item, the first media item comprising at least one of audio content and video content, determining a classification within an ontology of the subject matter of the first media item, using the ontology to infer other subject matter related to the determined subject matter of the first media item, and performing a search for other media items relevant to the determined subject matter of the first media item as a function of at least the other, related subject matter.
- the invention also provides An apparatus for identifying media items relevant to a first media item as a function of subject matter, the apparatus comprising: means for determining a subject matter of a first media item, the first media item comprising at least one of audio content and video content; means for classifying the first media item within an ontology as a function of its subject matter; means for analyzing the ontology to identify other, related subject matter to the subject matter of the selected media item; and means for performing a search for other media items relevant to the subject matter of the first media item as a function of at least the other, related subject matter according to the ontology.
- Figure 1 is a diagram illustrating a portion of an ontology in accordance with an embodiment of the present invention.
- Figure 2 is a diagram conceptually illustrating components of a system in accordance with an embodiment of the present invention.
- Figure 3 is a flow diagram illustrating operation in accordance with an embodiment of the present invention.
- Consumers of media content such as television programs, radio programs, videos, podcasts, digitally recorded music, and web pages will increasingly desire functionality for finding media content relevant to a particular interest of the consumer, particularly at the sub-asset level.
- Media refers to the forms in which content may be transmitted.
- audio e.g., music, speech
- visual photographs, drawings, motion pictures, web pages, animation
- These media are typically represented in electronic formats, such as, for example, HTTP, NNTP, UDP, JMS, TCP, MPEG, MP3, wave files, HTML, JPEG, TIFF, and PDF.
- transmitted media will likely involve other sensory data such as taste, smell and touch.
- media content providers and producers such as subscriber-based television network operators (e.g., cable, satellite and fiber optic television network operators), web site operators, podcasters, etc.
- subscriber-based television network operators e.g., cable, satellite and fiber optic television network operators
- web site operators e.g., web site operators
- podcasters etc.
- the media content e.g., television programs, radio programs, videos, digitally recorded music, podcasts, etc.
- an on-demand basis i.e., a consumer can consume any media items at any time of his or her choosing, rather than having to wait for a particular broadcast time.
- This already is the consumption paradigm for most web sites and podcasters.
- subscriber-based television networks already provide search and/or browse functions that allow their subscribers to search for media content.
- Video-On-Demand is a very popular service offered by many subscription television networks.
- Video-On-Demand is a service by which subscribers may choose programs from a menu for viewing at a time of each individual subscriber's choosing. A subscriber simply selects a program for viewing from a menu of programs that are available for viewing. The program, which is stored in memory at the headend or another server-side node of the network is then streamed to the subscriber's set top box immediately for viewing at that time.
- Media items are typically offered by programmers and network operators in generally predefined portions herein termed assets.
- assets For instance, television programs such as dramas, soap operas, reality shows, and sitcoms are typically broadcast in asset level units known as episodes that commonly are a half hour or an hour in length (including advertisements). Sporting events are broadcast in asset units of a single game.
- Music videos are commonly offered in asset units corresponding to a complete song or a complete concert performance.
- Consumers will increasingly consume media at the sub-asset level. That is, for instance, rather than watching an entire baseball game (a media asset), a consumer may watch only the parts where the team that he roots for is at bat or may only watch a highlight reel of the game (a sub-asset level segment). In another example, a viewer may view only the light saber fight scenes from the Star Wars movie series. Likewise, advertisers would be interested in buying advertising time within television content at the sub-asset level based on the subject matter of particular media segments.
- Content refers broadly to the information contained in the signal transmitted, and includes, for example, entertainment, news, and commercials.
- a media asset typically can conceptually be broken down into a plurality of segments at the sub-asset level, each having a cohesive subject or theme.
- the nature and duration of each segment will depend, of course, on the particular ontology used for purposes of segmentation as well as on the particular content of each program. For instance, most stage plays and motion pictures readily break down into two or three acts. Each such act can be a different segment.
- Television programs also can be segmented according to thematic elements. Certain programs, for instance, the television news magazine program 60 Minutes can readily be segmented into different news stories. Other programs, however, can be segmented based on more subtle thematic elements.
- a baseball game can be segmented by inning or at-bats, for instance.
- a typical James Bond movie can be segmented into a plurality of action segments, a plurality of dramatic segments, and a plurality romantic segments. The possibilities for segmentation based on thematic elements is virtually limitless and these are only the simplest of examples.
- consumers of media can search for media content of interest to them on the Internet through various search engines by entering a search string including terms that the consumer believes to be relevant to the type of subject matter for which he or she is searching.
- search functionality also is available in most subscriber-based television networks (e.g., cable television, fiber optic, and satellite based television networks) for searching for television programming.
- subscriber-based television networks e.g., cable television, fiber optic, and satellite based television networks
- search functionality is quite limited as compared to Internet search engines. For instance, some systems allow only literal title searching.
- search results often are not exactly what the consumer was seeking. This can be for several reasons. First, the consumer may simply have put in a poorly chosen search string of terms which returns results that are not relevant to the subject matter for which the consumer is looking. Second, a good search strategy may return results that are relevant to the subject matter of interest, but are too numerous to be useful to the consumer.
- a media content provider (be it a cable, satellite, or fiber optic television network operator, a Web site operator, a podcast provider, etc.) may wish to provide a feature to its users whereby a consumer can press a button while consuming particular media content and be presented with a user interface within which the user is presented a menu of other content available on the network (preferably at the segment level) having similar or related subject matter.
- a media content provider such as a cable, satellite, or fiber optic television network operator, a Web site operator, a podcast provider, etc.
- Patent Application No.12/343,790 filed 24 December 2008 entitled Identification of Segments Within Audio, Video, and Multimedia Items
- U.S. Patent Application No. 12/274,452 filed 20 November 2008 entitled Method and Apparatus for Delivering Video and Video-Related Content at Sub-Asset Level, all of which are owned by the same applicant as the present application, and which are incorporated herein fully by reference, discuss various aspects of one such system.
- U.S. Patent Application No. 12/274,452 particularly discusses an automated search function that can be offered to media consumers in the midst of consuming (e.g., viewing) one media asset (or segment thereof) that will search for other media items (assets, segments at the sub-asset level, or other items) that are pertinent to the subject matter of the media item being consumed.
- a user of an information network is offered supplemental content, the supplemental content being selected based at least partially on the subject matter of the media currently being consumed by the user.
- Information network refers to a collection of devices having a transport mechanism for exchanging information or content between the devices. Such networks may have any suitable architecture, including, for example, client-server, 3-tier architecture, N-tier architecture, distributed objects, loose coupling, or tight coupling.
- an exemplary embodiment of such a system might be implemented as part of the services offered to subscribers of a cable television network or as a feature on a website.
- a supplemental content search feature such as by depressing a dedicated button on a remote control unit or mouse clicking on an icon positioned somewhere on a computer monitor.
- the set top box STB
- the server performs a search for supplemental content that pertains to the particular media content being consumed by that consumer at that time and presents a list of such supplemental content to the viewer via a suitable user interface through which the viewer may select one or more for viewing.
- Aforementioned U.S. Patent Application No. 12/343,786 and U.S. Patent Application No. 12/343,790 collectively disclose techniques and apparatus for segmenting media items (such as media assets) into smaller segments (such as sub-assets), determining the boundaries and subject matter of contextually cohesive segments of the items, and classifying and organizing the segments for searching and browsing according to an ontology.
- media items such as media assets
- segments such as sub-assets
- U.S. Patent Application No. 12/343,790 provides a system for automatically identifying contextually cohesive segments, within media items (e.g., media assets) This task includes both identifying the beginnings and ends of such segments as well as the subject matter of the segments.
- searching for media content pertinent to a particular topic is an imperfect science.
- search engines search for content based on key words.
- aforementioned U.S. Patent Application Nos. 12/343,786 and 12/343,790 disclose suitable techniques for implementing such a system.
- One way of searching such content is to design a system that determines key words and other subject matter cues within the first media item using a number of techniques such as analyzing the closed caption stream, optical character recognition, video analytics, metadata analysis, etc., and then form a search string comprising those keywords for input into a search engine.
- the search engine may search for media content on the Internet and/or on a private network, such as a cable television network.
- a cable television service provider may search through its own database of television programming content (e.g., that has previously been analyzed for subject matter) for other media assets or segments in which those same keywords appear. The results might be weighed and ordered as a function of the number of times the keywords appear, the number of different keywords that appear and/or other criteria.
- the quality of the search results i.e., the pertinence of the results to the subject matter of the first media item, will depend on many factors, including, but not limited to, (1) the quality of the determination of the subject matter of the first media item, (2) the number of keywords that could be identified, (3) whether any of those keywords have double meanings, (4) the specificity of the keywords, and (5) the quantity of other media items having those keywords therein.
- the present invention offers a way to improve searches by capitalizing on a robust knowledge of subject matter in connection with specific knowledge domains.
- the database stores each segment's identity (such as by the identity of the media asset of which it forms a part and the time indexes within that asset of the start and end times of the segment) and its contextual information, e.g., in the form of a plurality of attribute/value pairs, a flat (table) database model of tuples (an ordered list of values), hierarchical data models, or relational data models).
- a flat (table) database model of tuples an ordered list of values
- hierarchical data models or relational data models.
- an ontology (or classification system) 103 is developed to provide a defined framework for classifying media segments by subject matter.
- An ontology essentially is a formal representation of a set of concepts within a domain and the relationships between those concepts. It is used to reason about the properties of that domain, and may be used to define the domain. Key elements of an ontology include:
- an interception is a turnover
- an interception may happen on a tipped pass
- An example of a restriction is “non-eligible receivers can catch a pass only if it is tipped by a defender”.
- An example of a rule is "plays involving ineligible receivers may be near interceptions" and, therefore, may be closely related to an interception.
- the segments are then indexed with respect to the ontology.
- the database can then be searched such as by a keyword search and/or the ontology can be examined for segments relevant to any desired subject matter (i.e., faceted searching).
- a keyword search e.g., a keyword search
- the ontology can be examined for segments relevant to any desired subject matter (i.e., faceted searching).
- preferably different portions of the ontology related to different knowledge domains are specifically designed as a function of those specific knowledge domains, thus making the ontology even more robust.
- a knowledge domain essentially is a high level theme or subject.
- knowledge domain refers to a relatively broad category of theme or subject, such as baseball, football, romance, Spanish, music, medicine, law, comedy.
- the breadth and subject of any knowledge domain within the ontology is entirely within the discretion of its creator. The only requirement is that a knowledge domain have sub-categories of subject matter.
- the present invention leverages the robust knowledge of subject matter inherently contained in the ontology to provide an improved way of locating media content relevant to a search or another piece of media content.
- the ontology and particularly, the knowledge domain specific portions of an ontology of a system such as disclosed in aforementioned US Patent Application No. 12/343,790 inherently have built-in to them a robust knowledge of various knowledge domains. Accordingly, a process of formulating query search strings incorporating the robust knowledge provided by the ontology is likely to substantially improve the quality of the search results, i.e., return more contextually pertinent results, in most cases.
- Indexing segments according to an ontology that has been developed with robust knowledge of the particular knowledge domain of interest would disclose the relatedness of the concept of an interception to the concepts of a fumble, a turnover, tipped pass, ineligible receiver, and a safety.
- designing a search engine that takes into account related concepts as well as the degree of relatedness in accordance with the structure of the ontology could provide much better results than a conventional search engine that does not take into account the robust knowledge inherent in the ontology, particularly the knowledge domain specific portions of the ontology.
- the ontology for football may include portions such as illustrated in Figure 1 .
- one portion of the ontology includes a category called “turnovers” and shows that it has subcategories “interception”, “fumble”, and “safety” (an “is a” relationship in the nature of a simple taxonomy).
- This type of relationship is visually represented in Figure 1 by solid lines and a tree structure.
- the ontology also reveals, for instance, that interception may happen on a tipped pass (the "may happen" relationship being represented by a dashed line).
- Many other relationship types, rules, restrictions and attributes also may be represented in the ontology.
- FIG. 2 is a block diagram illustrating conceptually the components of an exemplary system 200 incorporating the present invention.
- a collection of multimedia files (e.g., media assets) 201 exists that can be partitioned into coherent segments (e.g., at the sub-asset level) according to an ontology 202.
- the segments will be maintained in a segment database 203 that identifies the segments and their subject matter.
- the identification data for each segment may include, for instance, the identification of the asset of which it forms a part and the time indexes within the asset of the start and end times of the particular segment.
- the subject matter information may comprise virtually any information about the subject of the segment.
- the subject matter information in the segment database may be stored as one or more attribute/value pairs.
- one of the attributes may be "Key Offensive Players" and its value would be assigned the names (or other identification indicia) of the primary offense team players involved in the play.
- the ontology as well as the number of attributes and the specific attributes for any given segment can differ as a function of the particular knowledge domain of the asset from which the segment is taken. More specifically, just as mentioned above with respect to the ontology, the particular pieces of contextual information maintained in the database may be specific to the knowledge domain of the media items being segmented. Preferably, the specific knowledge domain is selected as a function of the knowledge domain of the media asset or item that is being segmented. For instance, the attributes stored in connection with a segment that forms part of a football competition may be different than the attributes that are stored for a segment that is part of a baseball competition, which are even further different than the attributes that are stored in connection with a segment that is part of a program about cooking.
- the knowledge domain of most media items is either known in advance of any subject matter analysis of the item (e.g., from the title of the asset) or is easily determinable via an initial subject matter analysis.
- the knowledge domain of the item may be input manually by a human operator. Alternately, it may be derived by analysis of the title of the asset. This can be done, for instance, by keyword analysis within the title or by comparing the title against a database of known program titles correlated to their knowledge domains.
- the specific pieces of information determined and stored with respect to a segment i.e., the attribute/value pairs stored in the segment database 203 also can be customized as a function of the specific knowledge domain of the item of which it forms a part (or the predicted interests of a particular demographic).
- the attributes for segments of a football competition may include Team On Offense, Team On Defense, Game Time, Down Number, Key Offensive Players, Key Defensive Players, Type of Play (e.g., kick off, point after attempt, punt regular play), Yards Gained/Lost, etc.
- the attributes for segments forming a portion of a baseball competition may be substantially different.
- the attributes that are to be stored in the database for a given segment may differ depending on the knowledge domain of the asset from which the segment is taken.
- Specialized attribute sets may be designed for the most common, relevant or popular knowledge domains for the given population of media assets to be segmented.
- a plurality 205 of different subject matter gathering processes 106-113 are utilized to determine the boundaries and subject matter of cohesive segments of the media assets 101.
- the process of identifying contextually cohesive segments of multimedia assets segmentation process 105 has at least two parts, namely, (1) identifying cohesive, meaningful segments within media items (e.g., identifying the beginning and end of a meaningful segment having a cohesive theme or subject) and (2) identifying that subject. Particularly, identifying keywords or other thematic elements in a multimedia file in order to identify subject matter is only half the battle. Delimiting the segments, i.e., determining the boundaries (beginning and end) of a cohesive segment is an additional complexity.
- segmenter 205 in Figure 2 may be utilized for determining the subject matter of media items, such as assets, and partitioning them into coherent segments as a function of their subject matter.
- software 206 is now available that can capture the closed caption stream within a media asset and analyze it for subject matter.
- software 207 is available that can analyze the audio portion of a multimedia stream and detect speech within the audio stream and convert the speech to text (which can further be analyzed for subject matter, just like the closed caption stream).
- voice recognition software can be used to detect the identity of a particular speaker within a media stream.
- certain types of multimedia files such as television programs of a particular title (e.g., "60 Minutes” or "Seinfeld") have a known set of individuals that are likely to speak during the program. In 60 Minutes, for instance, it would be the handful of reporters that regularly host segments of the program. In “Seinfeld”, it would be one of the handful of main characters - Jerry Seinfeld (played by actor Jerry Seinfeld), Elaine Benes played by actor Julia Louis-Dreyfus), Cosmo Kramer (played by actor Michael Richards), and George Costanza (played by actor Jason Alexander).
- Such software can be preprogrammed to recognize the voices of those main characters/actors and then used to recognize those voices to provide even richer subject matter data.
- audio analytics software 208 is now available that can analyze the non-speech aspects of the audio stream of an audio or multimedia file to determine additional subject matter information from sounds other than speech. For instance, such software can detect, recognize, and distinguish between, for instance, the sound of a crowd cheering or booing, sounds associated with being outdoors in a natural setting or being outdoors in an urban setting, or being indoors in a factory or an office or a residence, etc.
- U.S. Patent No. 7,177,881 discloses suitable software for detecting semantic events in an audio stream.
- optical character recognition software 209 can be used to determine text that appears in a scene. See, e.g. Li, Y. et al. "Reliable Video Clock Recognition," Pattern Recognition, 2006, 1 CPR 2006, 18th International Conference on Pattern Recognition .
- Such software can be used, for instance, to detect the clock in a timed sporting event. Specifically, knowledge of the game time could be useful in helping determine the nature of a scene. For instance, whether the clock is running or not could be informative as to whether the ball is in play during a football game. Furthermore, certain times during a sporting event are particularly important, such as two minutes before the end of a professional football game.
- optical character recognition can be used to determine the names of the actors, characters, and/or other significant persons in a television program or the like simply by reading the credits at the beginning and/or end of the program.
- video analytics software 210 is available that can analyze other visual content of a video or multimedia stream to determine subject matter information, e.g., indoors or outdoors, presence or absence of cars and other vehicles, presence or absence of human beings, presence or absence of non-human animals, etc.
- subject matter information e.g., indoors or outdoors, presence or absence of cars and other vehicles, presence or absence of human beings, presence or absence of non-human animals, etc.
- software is available today that can be used to actually recognize specific individuals by analyzing their faces.
- Metadata contained in a multimedia stream may be considered metadata, we here refer to additional information.
- the makers or distributors of television programs or third party providers sometimes insert metadata into the stream that might be useful in determining the subject matter of a program or of a portions of a program.
- metadata may include almost any relevant information, such as actors in a scene, timestamps identifying the beginnings and ends of various segments within a program, the names of the teams in a sporting event, the date and time that the sports event actually occurred, the number of the game within a complete season, etc.
- the segmenter 105 also may include software 111 for analyzing such metadata.
- the data may include, for instance, the time that each half inning commenced and ended, data for each at bat during the game, such as the identity of the batter, the result of the at-bat, the times at which the at-bat commenced and ended, the statistics of each player in the game, the score of the game at any given instance, the teams playing the game, etc.
- another software module 212 can be provided to analyze data obtained or otherwise obtained from external sources, such as Stats, Inc.
- optical character recognition (OCR) of the game clock in a sporting event also would be very useful in terms of aligning the game time with the media stream time.
- OCR optical character recognition
- external data available from sources such as Stats, Inc. includes data disclosing the time during the game that certain events (e.g., plays) occurred, but generally does not contain any information correlating the game time to the media stream time index.
- an alignment algorithm 121 for correlating game time with data stream time also would be a useful software component for purposes of identifying cohesive segments in connection with at least certain types of multimedia content, such as timed sports competitions.
- external data is widely available free of charge.
- additional subject matter information may be obtained via the Internet.
- Much information about sporting events and television shows is widely available on the Internet from any number of free sources.
- synopses of episodes of many television shows are widely available on the Internet, including character and actor lists, dates of first airing, episode numbers in the sequence of episodes, etc.).
- the present invention may rely on any or all of these techniques for determining the subject matter of a media item as well as the beginning and end of coherent segments corresponding to a particular subject matter.
- different subject matter information gathering processes for different knowledge domains may use different sets of these tools and/or use them in different ways or combinations.
- the same technologies in segmentor 105 may be used to determine the knowledge domains (i.e., the more general subject matter) of assets in embodiments in which such information is not predetermined so that the system can choose the particular set of technologies and particular attribute/value sets adapted to that knowledge domain for carrying out the segmentation.
- classification of media items need not be exclusive. For instance, a given segment may be properly assigned two or more relatively disparate contextual information within the ontology. For instance, a television program on the History Channel having a portion pertaining to the origination of the sport of golf in Scotland may be classified as pertaining to all of (1) history, (2) travel, and (3) sports.
- the segmentation also need not necessarily be discrete. Segments also may overlap. For instance, the same show on the History Channel mentioned above may start with a segment on Scottish history that evolves into a segment on the origins of golf and that even further evolves into a segment on Scottish dance music. Accordingly, a first segment may be defined as starting at timestamp 5 minutes:11 seconds in the program and ending at timestamp 9m:18s classified under History:European:Scotland. A second segment starting at 7m:39s and ending at 11m:52s may be classified under Sports:Solo:Golf and a third segment starting at 11m:13s and ending at 14m:09s may be classified under Music:Dance:Scottish. In this example, the various segments overlap each other in time.
- a segment can be any length, including zero (i.e., it is a single instant in time within the media item).
- the ones of the various information gathering processes used to analyze a particular media item and the manner of their use may be customized as a function of the knowledge domain of the particular item.
- the system operator may predetermine a plurality of subject matter information gathering processes, each adapted to a particularly relevant, popular, or common knowledge domain for assets within its collection of media assets and/or are popular interests among the expected users of the system (e.g., subscribers of a television service network employing the system or advertisers on that television network).
- a more generic, default information gathering process can be used for media items whose knowledge domain either cannot reasonably be determined or that do not fall into any of the other knowledge domain customized processes.
- the plurality of knowledge domains to which the ontology, subject matter information gathering processes, and/or attribute sets are customized should be specifically adapted for the types of media assets that commonly comprise television programming. For instance, the vast majority of network television programs fall in to one of a relatively small number of categories or knowledge domains.
- a specialized portion of the ontology, a specialized set of attributes, and/or a specialized subject matter information gathering process can be developed and used for each of these knowledge domains.
- the segments are then stored in the segment database 203 with all of their applicable attribute/value pairs.
- the media assets themselves do not necessarily need to be physically separated into distinct files at the segment level in database 203.
- the database 203 may merely comprise data identifying the segments.
- the segments also are indexed in accordance with the ontology 202.
- the overall ontology preferably is specifically adapted to classifying that type, e.g., multimedia items common to television programming.
- distinct portions of the ontology pertaining to different knowledge domains within television programming content may be specifically adapted to those knowledge domains.
- the ontology can now be used to determine the relevance of any media segment classified therein to any other media segment classified therein not merely by classification (e.g., an "is a" relationship), but by any number of relations.
- an improved search algorithm can be developed in connection with a particular subject matter by incorporating such related concepts and/or keywords in the searching algorithm.
- the searching algorithm may simply develop a search string (a set of words) to plug into a pre-existing search engine.
- the system will identify keywords derived from the subject matter of the media content using the tools discussed above.
- the ontology is consulted to identify related concepts and/or keywords. For instance, in the football example above, the ontology would disclose that an "interception” is a subcategory or child concept of the concept/keyword “turnover”. It also would disclose that sister forms of turnover to an "interception” include “fumbles" and "safeties”.
- the algorithm may also go down one level in the ontology to determine child concepts/keywords within the category of "interception” that may be useful search terms to add to the search string. Even further, the algorithm may look for other relations, rules, and/or restrictions, such as “interception” "may happen” on a “tipped pass,” etc.
- the various concepts/keywords developed through analysis of the ontology may be given different weights in terms of finding relevant documents. For instance, in a simple embodiment of this feature, those concepts/keywords not found directly through the subject matter analysis of the asset or segment being consumed, but identified as related concepts via the ontology may be given a lower weight in the search algorithm. In more complex embodiments, concepts from one level above may be weighed differently than sister concepts within the same level, which may be weighed differently than child concepts/keywords found by looking down one level in the ontology.
- the present invention is not limited to simply identifying related keywords for insertion into a search string.
- the present invention can be utilized to identify related concepts/keywords within the ontology and then incorporate those concepts/keywords in any type of searching algorithm, not just keyword searching.
- Another way to use the present invention in connection with a system such as the system disclosed in US patent application number 12/274,452 for automated searching for media content having subject matter similar to the subject matter of a particular media piece is to directly use the ontology to find other media content that is indexed within the ontology similarly to the indexing of the particular piece.
- FIG. 3 is a flowchart illustrating process flow in accordance with the principles of the present invention in an exemplary embodiment of an automated search feature such as the one disclosed in aforementioned US Patent Application No. 12/274,452 .
- an ontology is created specifically adapted to a specific type of media content.
- the overall ontology is adapted to its particular task, e.g., a subscription-based television service provider would adapt its ontology for multimedia content, and specifically the typical television programming type of multimedia content.
- a music service provider e.g., an online radio station, might use a very different ontology specifically adapted for audio content, and more specifically music content.
- the ontology preferably is developed with different portions thereof based on different knowledge domains, each particularly adapted to a different specific knowledge domain.
- the ontology can have within it different portions specifically adapted for sports games, cooking shows, situation comedies, reality shows, game shows, etc. If the larger domain is music, the specific knowledge domains might instead be rock, jazz, rhythm & blues and classical.
- a database is built storing the subject matter information for a plurality of media items. Furthermore, the ontology is populated with the media items, i.e., the various media items are classified within the ontology.
- the subject matter information and classification information for the library of media items can be collected, for instance, in the ways described hereinabove, including speech recognition analysis, OCR analysis, closed caption analysis, metadata analysis, audio analytics, video analytics, external data, etc.
- step 303 the media item that is currently being consumed is analyzed using any or all of the various aforementioned technologies to determine the subject matter of that media item.
- step 304 the item being consumed is classified within the ontology.
- step 305 the ontology is analyzed to determine additional concepts and/or keywords relevant to the subject matter information for the media item being consumed.
- step 306 a search is formulated for other media items having similar subject matter information to the item being consumed using the subject matter information collected directly from the media item as well as the related concepts and key words derived from the analysis of the ontology.
- step 307 the search is performed and the viewer is presented with the results of the search.
- the exemplary embodiments disclosed herein in connection with automated searching for related content to a media item being consumed by a viewer is merely exemplary.
- the concepts of the present invention can be applied in many other contexts. For instance, it can be applied to a search, such as an Internet search, developed by a human user when there is an available ontology relevant to the knowledge domain of the search.
- a search such as an Internet search
- the search terms or keywords input into a search engine by a human user can be run through that ontology to identify additional related terms according to the ontology and then an enhanced search string can be developed for use in performing the actual search (with or without making the human user aware of the modification).
- the ontology can be used to determine the relevance to each other of any two or more media segments already classified within the ontology. This embodiment might be useful in connection with tasks such as automatically generating playlists pertaining to a particular subject matter.
- an ontology can be used to help improve the parameters used to search for content relevant to any particular topic within the knowledge domain of the ontology in the complete absence of any content actually being indexed under the ontology.
- the existence of content indexed actually under the ontology would make it easier to locate and identify relevant content since the act of identifying other related concepts within the ontology would inherently also identify the corresponding content indexed under those related concepts in the ontology.
- all of the media items 101 are stored in a digital memory as digital files.
- the ontology and the segment database also are stored in a computer or other digital memory.
- the various subject matter information gathering modules are preferably implemented as software routines running on a general or special purpose digital processing device. However, the processes also could be implemented in any of a number of other reasonable manners, including, but not limited to, integrated circuits, combinational logic circuits, field programmable gate arrays, analog circuitry, microprocessors, state machines, and/or combinations of any of the above.
- the mechanisms for formulating the search strategy as well as the mechanism for performing the search also are preferably implemented as software routines, but also could be implemented in any of the aforementioned manners.
- the subject matter information gathering process and/or the attribute/value pairs for the segment database particularly for a plurality of different specific knowledge domains based on a robust knowledge of each such knowledge domain (e.g., cooking, football, sitcoms, law dramas), one can provide much richer and more robust search and retrieval functionality for users.
- a robust knowledge of each such knowledge domain e.g., cooking, football, sitcoms, law dramas
- the ontology 105 can be continuously refined as types of programming, products, services, demographics, etc. are developed or themselves become more refined.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Library & Information Science (AREA)
- Computational Linguistics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention pertains to methods, systems, and apparatus for identifying media items relevant to a selected subject matter, the method comprising determining the subject matter of a first media item, the first media item comprising at least one of audio content and video content, determining the classification within an ontology of the subject matter of the first media item, analyzing the ontology to identify other subject matter related to the subject matter of the first media item, and performing a search for other media items relevant to the subject matter of the first media item as a function of at least the other related subject matter according to the ontology.
Description
- The invention pertains to a process and apparatus for organizing segments of audio, video, and media files and determining the relevance of such segments to each other or to a query.
- Until fairly recently, individuals consumed audio, video, and other media content in relatively few forms (television, movies, musical albums) from relatively few sources (television stations, movie theaters, radio stations, CDs). However, with the advent of the Internet and an explosion in the availability of low cost electronic consumer products, the forms and potential sources of such content have become much more numerous. Today, individuals can consume such content on computers at home via the internet, on any number of portable devices with memory for storing content, on mobile devices with wireless network connectivity to content, on televisions, in movie theaters, etc. Furthermore, the potential sources of audio, video, and multimedia content are virtually limitless. For instance, subscription-based television network systems, such as cable television, now provide video on demand offering in addition to standard broadcast television. They also allow subscribers to record broadcast television programs and watch them at a time of their own choosing and with the ability to control the content stream, such as by fast forward, skip, pause, rewind, etc.
- Even further, almost anyone with a computer can now create and widely publish their own audio, video, and multimedia content on the Internet through such outlets as podcasts, videos published via websites such as myspace.com or youtube.com. Accordingly, both the amount of available content and the specificity of the content has increased dramatically.
- As both the volume and specificity of audio, video, and media content increase, it is expected that consumers will increasingly consume such content, including television programs, movies, music videos, podcasts, musical albums, and other audio, video, and multimedia assets at the sub-asset level. That is, for instance, rather than watching an entire baseball game, a consumer may watch only the parts where the team that he roots for is at bat or may only watch a highlight reel of the game. In another example, a viewer may view only the light saber fight scenes from the Star Wars movie series. In yet other examples, a viewer may watch only the sports segment or the weather segment of the evening news program or listen to only a single song from a CD or album.
- Presently, the only way a consumer of media content can access a segment of particular interest to that consumer within a media asset is to scan through the asset in a linear fashion, such as by using a fast-forward or rewind function of a media player, to find the desired content.
- "Media" refers to the forms in which content may be transmitted. Presently, the most common transmitted media are audio (e.g., music, speech) and visual (photographs, drawings, motion pictures, web pages, animation). These media are typically represented in electronic formats, such as, for example, HTTP, NNTP, UDP, JMS, TCP, MPEG, MP3, wave files, HTML, JPEG, TIFF, and PDF. As transmission technologies become more advanced, however, transmitted media will likely involve other sensory data such as taste, smell and touch..
- The decision as to which segments within a complete media item any individual wishes to view, of course, is based on the subject matter of the content of the segment, hereinafter termed contextual information or subject matter. "Contextual information" or "subject matter" refers broadly to the topic or theme of the content and can be virtually anything within the realm of human knowledge, such as baseball, strike out, fast ball, stolen base, mountains, scary, happy, George Carlin, nighttime, cool, winner. The nature and duration of each segment will depend, of course, on the particular ontology.
- Furthermore, as is well-known, advertisers often purchase advertisement time or space within media assets such as television programs, web pages, podcasts, and radio programs based on the subject matter of the media. Specifically, advertisers commonly are interested in a particular demographic of media consumers that can range from the very broad to the extremely narrow. For instance, a producer of beer might be interested in a demographic of male media consumers aged 18-45, whereas a producer of anti-aging face cream for women might be interested in a demographic comprising female viewers aged 30-70. The subject matter of a media asset often has a very high correlation to a specific demographic. Therefore, the producer of anti-aging face cream may be much more interested in placing its advertisement in the middle of a soap opera rather than a football competition because the soap opera will be viewed by many more individuals within the demographic that is likely to buy its product than the football competition, even if the football competition has a much larger overall viewing audience than the soap opera.
- Thus, not only do individuals expend a significant amount of effort selecting which media assets they consume, but a great deal of effort is expended by media content providers, (e.g., individual television and radio stations, cable, fiber optic and satellite subscription-based television network operators, internet service providers), media content producers (e.g., television and radio program producers, podcasters, website operators) and advertisers in determining what subject matters of such media appeal to particular demographics for advertisement placement and other purposes.
- The invention pertains to methods, systems, and apparatus for identifying media items relevant to a subject matter of a first media item, the method comprising determining the subject matter of a first media item, the first media item comprising at least one of audio content and video content, determining a classification within an ontology of the subject matter of the first media item, using the ontology to infer other subject matter related to the determined subject matter of the first media item, and performing a search for other media items relevant to the determined subject matter of the first media item as a function of at least the other, related subject matter.
- From another aspect the invention also provides An apparatus for identifying media items relevant to a first media item as a function of subject matter, the apparatus comprising: means for determining a subject matter of a first media item, the first media item comprising at least one of audio content and video content; means for classifying the first media item within an ontology as a function of its subject matter; means for analyzing the ontology to identify other, related subject matter to the subject matter of the selected media item; and means for performing a search for other media items relevant to the subject matter of the first media item as a function of at least the other, related subject matter according to the ontology.
-
Figure 1 is a diagram illustrating a portion of an ontology in accordance with an embodiment of the present invention. -
Figure 2 is a diagram conceptually illustrating components of a system in accordance with an embodiment of the present invention. -
Figure 3 is a flow diagram illustrating operation in accordance with an embodiment of the present invention. - Consumers of media content such as television programs, radio programs, videos, podcasts, digitally recorded music, and web pages will increasingly desire functionality for finding media content relevant to a particular interest of the consumer, particularly at the sub-asset level.
- "Media" refers to the forms in which content may be transmitted. Presently, the most common transmitted media are audio (e.g., music, speech) and visual (photographs, drawings, motion pictures, web pages, animation). These media are typically represented in electronic formats, such as, for example, HTTP, NNTP, UDP, JMS, TCP, MPEG, MP3, wave files, HTML, JPEG, TIFF, and PDF. As transmission technologies become more advanced, however, transmitted media will likely involve other sensory data such as taste, smell and touch..
- As an example, it is envisioned that media content providers and producers, such as subscriber-based television network operators (e.g., cable, satellite and fiber optic television network operators), web site operators, podcasters, etc., eventually will offer all or most of the media content (e.g., television programs, radio programs, videos, digitally recorded music, podcasts, etc.), to consumers on an on-demand basis (i.e., a consumer can consume any media items at any time of his or her choosing, rather than having to wait for a particular broadcast time). This already is the consumption paradigm for most web sites and podcasters. Furthermore, many subscriber-based television networks already provide search and/or browse functions that allow their subscribers to search for media content. For instance, Video-On-Demand (VOD) is a very popular service offered by many subscription television networks. Video-On-Demand is a service by which subscribers may choose programs from a menu for viewing at a time of each individual subscriber's choosing. A subscriber simply selects a program for viewing from a menu of programs that are available for viewing. The program, which is stored in memory at the headend or another server-side node of the network is then streamed to the subscriber's set top box immediately for viewing at that time.
- Media items are typically offered by programmers and network operators in generally predefined portions herein termed assets. For instance, television programs such as dramas, soap operas, reality shows, and sitcoms are typically broadcast in asset level units known as episodes that commonly are a half hour or an hour in length (including advertisements). Sporting events are broadcast in asset units of a single game. Music videos are commonly offered in asset units corresponding to a complete song or a complete concert performance.
- In the television arts, professionals on the business side of the art tend to refer to these as "assets," whereas professionals on the research and technical side of the art more often refer to them as "documents." In either event, the concept of a media "asset" or "document" is well understood in the industry as well as among content consumers (who may not necessarily know the term "document" or "asset," but know the concept). For instance, a typical television guide printed in a newspaper or the electronic program guides commonly provided by a subscriber-based television network are well known to virtually all television viewers and generally list multimedia content at the asset level.
- As both the volume and specificity of media content increases, it is expected that consumers will increasingly consume media at the sub-asset level. That is, for instance, rather than watching an entire baseball game (a media asset), a consumer may watch only the parts where the team that he roots for is at bat or may only watch a highlight reel of the game (a sub-asset level segment). In another example, a viewer may view only the light saber fight scenes from the Star Wars movie series. Likewise, advertisers would be interested in buying advertising time within television content at the sub-asset level based on the subject matter of particular media segments. "Content" refers broadly to the information contained in the signal transmitted, and includes, for example, entertainment, news, and commercials.
- A media asset typically can conceptually be broken down into a plurality of segments at the sub-asset level, each having a cohesive subject or theme. The nature and duration of each segment will depend, of course, on the particular ontology used for purposes of segmentation as well as on the particular content of each program. For instance, most stage plays and motion pictures readily break down into two or three acts. Each such act can be a different segment. Television programs also can be segmented according to thematic elements. Certain programs, for instance, the television news magazine program 60 Minutes can readily be segmented into different news stories. Other programs, however, can be segmented based on more subtle thematic elements. A baseball game can be segmented by inning or at-bats, for instance. A typical James Bond movie can be segmented into a plurality of action segments, a plurality of dramatic segments, and a plurality romantic segments. The possibilities for segmentation based on thematic elements is virtually limitless and these are only the simplest of examples.
- Presently, consumers of media can search for media content of interest to them on the Internet through various search engines by entering a search string including terms that the consumer believes to be relevant to the type of subject matter for which he or she is searching. Such functionality also is available in most subscriber-based television networks (e.g., cable television, fiber optic, and satellite based television networks) for searching for television programming. However, in many systems, the search functionality is quite limited as compared to Internet search engines. For instance, some systems allow only literal title searching.
- Even with a robust Internet search engine, the search results often are not exactly what the consumer was seeking. This can be for several reasons. First, the consumer may simply have put in a poorly chosen search string of terms which returns results that are not relevant to the subject matter for which the consumer is looking. Second, a good search strategy may return results that are relevant to the subject matter of interest, but are too numerous to be useful to the consumer.
- Furthermore, automatic initiation and/or formulation of searches for content relevant to a particular subject matter will become increasingly common in the future. Particularly, a media content provider (be it a cable, satellite, or fiber optic television network operator, a Web site operator, a podcast provider, etc.) may wish to provide a feature to its users whereby a consumer can press a button while consuming particular media content and be presented with a user interface within which the user is presented a menu of other content available on the network (preferably at the segment level) having similar or related subject matter. For instance,
U.S. Patent Application No. 12/343,786, filed 24 December 2008 U.S. Patent Application No.12/343,790, filed 24 December 2008 U.S. Patent Application No. 12/274,452, filed 20 November 2008 - For instance, above noted
U.S. Patent Application No. 12/274,452 particularly discusses an automated search function that can be offered to media consumers in the midst of consuming (e.g., viewing) one media asset (or segment thereof) that will search for other media items (assets, segments at the sub-asset level, or other items) that are pertinent to the subject matter of the media item being consumed. More particularly, a user of an information network is offered supplemental content, the supplemental content being selected based at least partially on the subject matter of the media currently being consumed by the user. "Information network" refers to a collection of devices having a transport mechanism for exchanging information or content between the devices. Such networks may have any suitable architecture, including, for example, client-server, 3-tier architecture, N-tier architecture, distributed objects, loose coupling, or tight coupling. - For instance, an exemplary embodiment of such a system might be implemented as part of the services offered to subscribers of a cable television network or as a feature on a website. Let us consider as an example an individual consumer who is watching a particular television program, in this example, a major league baseball game between the Philadelphia Phillies and the New York Mets. The consumer is permitted at any time during the program to activate a supplemental content search feature, such as by depressing a dedicated button on a remote control unit or mouse clicking on an icon positioned somewhere on a computer monitor. When the feature is thus selected, the set top box (STB), for instance, sends a signal upstream to a server requesting invocation of the feature. In response, the server performs a search for supplemental content that pertains to the particular media content being consumed by that consumer at that time and presents a list of such supplemental content to the viewer via a suitable user interface through which the viewer may select one or more for viewing.
- Aforementioned
U.S. Patent Application No. 12/343,786 andU.S. Patent Application No. 12/343,790 collectively disclose techniques and apparatus for segmenting media items (such as media assets) into smaller segments (such as sub-assets), determining the boundaries and subject matter of contextually cohesive segments of the items, and classifying and organizing the segments for searching and browsing according to an ontology. The context being referred to in the terms "contextual information" and "contextually cohesive" is the ontology within which the subject matter of the media content is being classified. - More particularly,
U.S. Patent Application No. 12/343,790 provides a system for automatically identifying contextually cohesive segments, within media items (e.g., media assets) This task includes both identifying the beginnings and ends of such segments as well as the subject matter of the segments. - As noted above, searching for media content pertinent to a particular topic, be it a search query manually generated by a user or an automated search for other media items related to the subject matter of a first media item, such as disclosed in aforementioned
U.S. Patent Application No. 12/274,452 , is an imperfect science. Typically, search engines search for content based on key words. In the latter example of an automated search for additional media items pertinent to the subject matter of a first media item, aforementionedU.S. Patent Application Nos. 12/343,786 and12/343,790 - The quality of the search results, i.e., the pertinence of the results to the subject matter of the first media item, will depend on many factors, including, but not limited to, (1) the quality of the determination of the subject matter of the first media item, (2) the number of keywords that could be identified, (3) whether any of those keywords have double meanings, (4) the specificity of the keywords, and (5) the quantity of other media items having those keywords therein.
- As previously noted, some or all of the results of such a search may not be particularly pertinent to the consumer's interests.
- The mere matching of keywords often will not find all of the relevant content or the most relevant content. Particularly, depending on the particular user's interests, segments having very different keywords actually may be closely related depending on the user's viewpoint. The inverse also is true. That is, the identical word may have very different and unrelated meanings in different contexts.
- Let us consider an example to illustrate the concept. Suppose a viewer is watching a football game, and particularly the Super Bowl game of 2008 between the New England Patriots and the New York Giants. Let us also assume that a play occurs in which the quarterback for the New England Patriots, Eli Manning, throws an interception. During or immediately after this play, the viewer activates the automated search function with the hope of viewing other plays in which the New England Patriots offensive team turned over the ball to the other team. As anyone with a robust knowledge of the game of football will know, an interception is only one of several ways in which a turnover can occur in a football game. Another way is a fumble. Yet another way is a safety. The above example illustrates an "is a" relationship in the nature of a simple classification system or taxonomy. However, many other important relationships between concepts can be represented in a robust ontology. Such relationship may include "belongs to," "is a part of," and "may include." For instance, a tipped pass often leads to an interception or a near interception (a pass that almost was intercepted). It also allows otherwise ineligible receivers to receive a pass. Accordingly, plays in which a tipped pass occurs and/or an ineligible receiver catches a pass may be a near interception and, therefore, highly relevant to another play that is an interception. Thus, as a practical example, it is quite likely that a contextual analysis of the play in which the interception occurred may yield the keywords "New England Patriots," "Eli Manning," and "interception." A search using these keywords is likely to miss turnovers that occurred as the result of a fumble or safety and near interceptions.
- As can be seen from this example, a robust knowledge of football would enable the formulation of a better search for pertinent content. Thus, if the viewer had a robust knowledge of football and entered his own search string, he may have thought of adding the words "fumble," "safety," "tipped pass" and/or "ineligible receiver" to the search string or of including the term "turnover" in addition to or instead of "interception."
- The present invention offers a way to improve searches by capitalizing on a robust knowledge of subject matter in connection with specific knowledge domains.
- Particularly, aforementioned
U.S. Patent Applications 12/274,452 ,12/343,786 12/343,790 - In order to develop such a database 101, an ontology (or classification system) 103 is developed to provide a defined framework for classifying media segments by subject matter. An ontology essentially is a formal representation of a set of concepts within a domain and the relationships between those concepts. It is used to reason about the properties of that domain, and may be used to define the domain. Key elements of an ontology include:
- Classes: sets, collections, concepts, types of objects, or kinds of things
- Attributes: aspects, properties, features, characteristics, or parameters that objects (and classes) can have
- Relations: ways in which classes and individuals can be related to one another
- Restrictions: formally stated descriptions of what must be true in order for some assertion to be accepted as input
- Rules: statements in the form of an if-then (antecedent-consequent) sentence that describe the logical inferences that can be drawn from an assertion in a particular form
- Thus, for instance, "an interception is a turnover" is a relationship. Also, "an interception may happen on a tipped pass" also is a relationship. An example of a restriction is "non-eligible receivers can catch a pass only if it is tipped by a defender". An example of a rule is "plays involving ineligible receivers may be near interceptions" and, therefore, may be closely related to an interception.
- The segments are then indexed with respect to the ontology. The database can then be searched such as by a keyword search and/or the ontology can be examined for segments relevant to any desired subject matter (i.e., faceted searching). Furthermore, as discussed in detail in aforementioned
U.S. Patent Application No. 12/343,790 , preferably different portions of the ontology related to different knowledge domains are specifically designed as a function of those specific knowledge domains, thus making the ontology even more robust. - A knowledge domain essentially is a high level theme or subject. As used herein, "knowledge domain" refers to a relatively broad category of theme or subject, such as baseball, football, romance, Spanish, music, medicine, law, comedy. The breadth and subject of any knowledge domain within the ontology is entirely within the discretion of its creator. The only requirement is that a knowledge domain have sub-categories of subject matter.
- The present invention leverages the robust knowledge of subject matter inherently contained in the ontology to provide an improved way of locating media content relevant to a search or another piece of media content. Particularly, the ontology, and particularly, the knowledge domain specific portions of an ontology of a system such as disclosed in aforementioned
US Patent Application No. 12/343,790 - Indexing segments according to an ontology that has been developed with robust knowledge of the particular knowledge domain of interest, (e.g., football) would disclose the relatedness of the concept of an interception to the concepts of a fumble, a turnover, tipped pass, ineligible receiver, and a safety. Thus, designing a search engine that takes into account related concepts as well as the degree of relatedness in accordance with the structure of the ontology could provide much better results than a conventional search engine that does not take into account the robust knowledge inherent in the ontology, particularly the knowledge domain specific portions of the ontology.
- Thus, for instance, in the example above, the ontology for football may include portions such as illustrated in
Figure 1 . As can be seen, one portion of the ontology includes a category called "turnovers" and shows that it has subcategories "interception", "fumble", and "safety" (an "is a" relationship in the nature of a simple taxonomy). This type of relationship is visually represented inFigure 1 by solid lines and a tree structure. The ontology also reveals, for instance, that interception may happen on a tipped pass (the "may happen" relationship being represented by a dashed line). Many other relationship types, rules, restrictions and attributes also may be represented in the ontology. Thus, if a subject matter analysis of the play yields only the word "interception" as a keyword, consultation of the ontology discloses that concept of an "interception" is a type of "turnover" and that the concepts of a "fumble" and a "safety" are sister concepts to the concept of an "interception" because all three are forms of a "turnover." Thus, the search query can be modified to include one or more of "turnover", "fumble", and "safety" as keywords in addition to "interception." The terms "fumble" and "safety" may be weighed lower than the terms "interception" and/or "turnover" so that interceptions are weighted more heavily than fumbles and safeties (since they, obviously, are more closely related to the play in question than a fumble or a safety). -
Figure 2 is a block diagram illustrating conceptually the components of an exemplary system 200 incorporating the present invention. A collection of multimedia files (e.g., media assets) 201 exists that can be partitioned into coherent segments (e.g., at the sub-asset level) according to anontology 202. The segments will be maintained in asegment database 203 that identifies the segments and their subject matter. The identification data for each segment may include, for instance, the identification of the asset of which it forms a part and the time indexes within the asset of the start and end times of the particular segment. The subject matter information may comprise virtually any information about the subject of the segment. The subject matter information in the segment database may be stored as one or more attribute/value pairs. Thus, using as an example, a segment comprising a single play in a football competition, one of the attributes may be "Key Offensive Players" and its value would be assigned the names (or other identification indicia) of the primary offense team players involved in the play. - The ontology as well as the number of attributes and the specific attributes for any given segment can differ as a function of the particular knowledge domain of the asset from which the segment is taken. More specifically, just as mentioned above with respect to the ontology, the particular pieces of contextual information maintained in the database may be specific to the knowledge domain of the media items being segmented. Preferably, the specific knowledge domain is selected as a function of the knowledge domain of the media asset or item that is being segmented. For instance, the attributes stored in connection with a segment that forms part of a football competition may be different than the attributes that are stored for a segment that is part of a baseball competition, which are even further different than the attributes that are stored in connection with a segment that is part of a program about cooking.
- Generally, the knowledge domain of most media items is either known in advance of any subject matter analysis of the item (e.g., from the title of the asset) or is easily determinable via an initial subject matter analysis. The knowledge domain of the item may be input manually by a human operator. Alternately, it may be derived by analysis of the title of the asset. This can be done, for instance, by keyword analysis within the title or by comparing the title against a database of known program titles correlated to their knowledge domains. In any event, once the knowledge domain of the media item is determined (e.g., football, baseball, sitcom, reality show, reality competition, game show, etc.), the specific pieces of information determined and stored with respect to a segment (i.e., the attribute/value pairs stored in the segment database 203) also can be customized as a function of the specific knowledge domain of the item of which it forms a part (or the predicted interests of a particular demographic).
- Thus, for instance, continuing with the football competition example, the attributes for segments of a football competition may include Team On Offense, Team On Defense, Game Time, Down Number, Key Offensive Players, Key Defensive Players, Type of Play (e.g., kick off, point after attempt, punt regular play), Yards Gained/Lost, etc.
- On the other hand, the attributes for segments forming a portion of a baseball competition may be substantially different.
- In short, the attributes that are to be stored in the database for a given segment may differ depending on the knowledge domain of the asset from which the segment is taken. Specialized attribute sets may be designed for the most common, relevant or popular knowledge domains for the given population of media assets to be segmented.
- Thus, in a preferred embodiment of the invention, a
plurality 205 of different subject matter gathering processes 106-113 are utilized to determine the boundaries and subject matter of cohesive segments of the media assets 101. - The process of identifying contextually cohesive segments of multimedia assets segmentation process 105 has at least two parts, namely, (1) identifying cohesive, meaningful segments within media items (e.g., identifying the beginning and end of a meaningful segment having a cohesive theme or subject) and (2) identifying that subject. Particularly, identifying keywords or other thematic elements in a multimedia file in order to identify subject matter is only half the battle. Delimiting the segments, i.e., determining the boundaries (beginning and end) of a cohesive segment is an additional complexity.
- Various technologies, generally represented within
segmenter 205 inFigure 2 may be utilized for determining the subject matter of media items, such as assets, and partitioning them into coherent segments as a function of their subject matter. - Many technologies are available now that can be adapted for use for identifying media segments either as stand-alone components or in combination within the present invention. For instance,
software 206 is now available that can capture the closed caption stream within a media asset and analyze it for subject matter. Further,software 207 is available that can analyze the audio portion of a multimedia stream and detect speech within the audio stream and convert the speech to text (which can further be analyzed for subject matter, just like the closed caption stream). - In fact, voice recognition software can be used to detect the identity of a particular speaker within a media stream. For instance, certain types of multimedia files, such as television programs of a particular title (e.g., "60 Minutes" or "Seinfeld") have a known set of individuals that are likely to speak during the program. In 60 Minutes, for instance, it would be the handful of reporters that regularly host segments of the program. In "Seinfeld", it would be one of the handful of main characters - Jerry Seinfeld (played by actor Jerry Seinfeld), Elaine Benes played by actor Julia Louis-Dreyfus), Cosmo Kramer (played by actor Michael Richards), and George Costanza (played by actor Jason Alexander). Such software can be preprogrammed to recognize the voices of those main characters/actors and then used to recognize those voices to provide even richer subject matter data.
- Additionally,
audio analytics software 208 is now available that can analyze the non-speech aspects of the audio stream of an audio or multimedia file to determine additional subject matter information from sounds other than speech. For instance, such software can detect, recognize, and distinguish between, for instance, the sound of a crowd cheering or booing, sounds associated with being outdoors in a natural setting or being outdoors in an urban setting, or being indoors in a factory or an office or a residence, etc. For example,U.S. Patent No. 7,177,881 discloses suitable software for detecting semantic events in an audio stream. - Even further, optical
character recognition software 209 can be used to determine text that appears in a scene. See, e.g. Li, Y. et al. "Reliable Video Clock Recognition," Pattern Recognition, 2006, 1 CPR 2006, 18th International Conference on Pattern Recognition. Such software can be used, for instance, to detect the clock in a timed sporting event. Specifically, knowledge of the game time could be useful in helping determine the nature of a scene. For instance, whether the clock is running or not could be informative as to whether the ball is in play during a football game. Furthermore, certain times during a sporting event are particularly important, such as two minutes before the end of a professional football game. Likewise, optical character recognition can be used to determine the names of the actors, characters, and/or other significant persons in a television program or the like simply by reading the credits at the beginning and/or end of the program. - Furthermore,
video analytics software 210 is available that can analyze other visual content of a video or multimedia stream to determine subject matter information, e.g., indoors or outdoors, presence or absence of cars and other vehicles, presence or absence of human beings, presence or absence of non-human animals, etc. In fact, software is available today that can be used to actually recognize specific individuals by analyzing their faces. - Even further, there may be significant metadata contained in a multimedia stream. While a closed captioning stream may be considered metadata, we here refer to additional information. Particularly, the makers or distributors of television programs or third party providers sometimes insert metadata into the stream that might be useful in determining the subject matter of a program or of a portions of a program. Such metadata may include almost any relevant information, such as actors in a scene, timestamps identifying the beginnings and ends of various segments within a program, the names of the teams in a sporting event, the date and time that the sports event actually occurred, the number of the game within a complete season, etc. Accordingly, the segmenter 105 also may include software 111 for analyzing such metadata.
- Even further, companies now exist that provide the services of generating and selling data about sporting events, television programs, and other events. For instance, Stats, Inc. of Northbrook, IL, USA sells such metadata about sporting events. Thus, taking a baseball game as an example, the data may include, for instance, the time that each half inning commenced and ended, data for each at bat during the game, such as the identity of the batter, the result of the at-bat, the times at which the at-bat commenced and ended, the statistics of each player in the game, the score of the game at any given instance, the teams playing the game, etc. Accordingly, another
software module 212 can be provided to analyze data obtained or otherwise obtained from external sources, such as Stats, Inc. - Furthermore, the aforementioned optical character recognition (OCR) of the game clock in a sporting event also would be very useful in terms of aligning the game time with the media stream time. For instance, external data available from sources such as Stats, Inc. includes data disclosing the time during the game that certain events (e.g., plays) occurred, but generally does not contain any information correlating the game time to the media stream time index. Thus, an alignment algorithm 121 for correlating game time with data stream time also would be a useful software component for purposes of identifying cohesive segments in connection with at least certain types of multimedia content, such as timed sports competitions.
- Furthermore, external data is widely available free of charge. For instance, additional subject matter information may be obtained via the Internet. Particularly, much information about sporting events and television shows is widely available on the Internet from any number of free sources. For instance, synopses of episodes of many television shows are widely available on the Internet, including character and actor lists, dates of first airing, episode numbers in the sequence of episodes, etc.).
- The present invention may rely on any or all of these techniques for determining the subject matter of a media item as well as the beginning and end of coherent segments corresponding to a particular subject matter. Also, as previously noted, different subject matter information gathering processes for different knowledge domains may use different sets of these tools and/or use them in different ways or combinations. Furthermore, as previously mentioned, the same technologies in segmentor 105 may be used to determine the knowledge domains (i.e., the more general subject matter) of assets in embodiments in which such information is not predetermined so that the system can choose the particular set of technologies and particular attribute/value sets adapted to that knowledge domain for carrying out the segmentation.
- It should be noted that the classification of media items need not be exclusive. For instance, a given segment may be properly assigned two or more relatively disparate contextual information within the ontology. For instance, a television program on the History Channel having a portion pertaining to the origination of the sport of golf in Scotland may be classified as pertaining to all of (1) history, (2) travel, and (3) sports.
- It should be understood, that the example above is simplified for purposes of illustrating the proposition being discussed. In actuality, of course, a segment about the history and origins of golf in Scotland would be classified and sub-classified to multiple levels according to an ontology. For instance, in a robust ontology, this segment would not be merely classified under history, but probably would be further sub-classified under European history, and even further sub-classified under Scottish history, etc. It would further be classified not merely under travel, but probably under travel, then sub-classified under European travel, and then even further sub-classified under Scottish travel, etc. Finally, it also would not merely be classified under sports, but, for instance, under sports and further sub-classified under solo sports, and even further sub-classified under golf.
- The segmentation also need not necessarily be discrete. Segments also may overlap. For instance, the same show on the History Channel mentioned above may start with a segment on Scottish history that evolves into a segment on the origins of golf and that even further evolves into a segment on Scottish dance music. Accordingly, a first segment may be defined as starting at timestamp 5 minutes:11 seconds in the program and ending at timestamp 9m:18s classified under History:European:Scotland. A second segment starting at 7m:39s and ending at 11m:52s may be classified under Sports:Solo:Golf and a third segment starting at 11m:13s and ending at 14m:09s may be classified under Music:Dance:Scottish. In this example, the various segments overlap each other in time.
- Even further, a segment can be any length, including zero (i.e., it is a single instant in time within the media item).
- The ones of the various information gathering processes used to analyze a particular media item and the manner of their use may be customized as a function of the knowledge domain of the particular item. The system operator may predetermine a plurality of subject matter information gathering processes, each adapted to a particularly relevant, popular, or common knowledge domain for assets within its collection of media assets and/or are popular interests among the expected users of the system (e.g., subscribers of a television service network employing the system or advertisers on that television network). A more generic, default information gathering process can be used for media items whose knowledge domain either cannot reasonably be determined or that do not fall into any of the other knowledge domain customized processes.
- For instance, if the present invention is to be implemented on a subscription-based television service network, then the plurality of knowledge domains to which the ontology, subject matter information gathering processes, and/or attribute sets are customized should be specifically adapted for the types of media assets that commonly comprise television programming. For instance, the vast majority of network television programs fall in to one of a relatively small number of categories or knowledge domains. For instance, probably the vast majority of programs made for television fall into one of the following domains: news and current events, situational comedies, law-based dramas, police-based dramas, medical-based dramas, reality TV, reality competitions, sports competitions (which might further be broken down into a handful of the most popular sports, such as football, hockey, baseball, basketball, soccer, golf), children's cartoons, daytime soap operas, educational or informative (history, travel, technology), sketch comedy, talk shows, and game shows.
- Hence, a specialized portion of the ontology, a specialized set of attributes, and/or a specialized subject matter information gathering process can be developed and used for each of these knowledge domains.
- Once the segments are determined and the subject matter information has been gathered, the segments are then stored in the
segment database 203 with all of their applicable attribute/value pairs. - It should be understood that the media assets themselves do not necessarily need to be physically separated into distinct files at the segment level in
database 203. For instance, thedatabase 203 may merely comprise data identifying the segments. - The segments also are indexed in accordance with the
ontology 202. Again, if the overall system is to be used for a specific type of media, e.g., a television programming, then the overall ontology preferably is specifically adapted to classifying that type, e.g., multimedia items common to television programming. Furthermore, as previously noted, distinct portions of the ontology pertaining to different knowledge domains within television programming content may be specifically adapted to those knowledge domains. - In addition to simply populating the
segment database 203 with the data for a plurality of media segments, those segments also are indexed under the ontology, i.e., the ontology is populated with the segments, as illustrated inoval 215. Finally, as illustrated byoval 216, the ontology can now be used to determine the relevance of any media segment classified therein to any other media segment classified therein not merely by classification (e.g., an "is a" relationship), but by any number of relations. - Using the ontology to identify related concepts and keywords, an improved search algorithm can be developed in connection with a particular subject matter by incorporating such related concepts and/or keywords in the searching algorithm.
- In accordance with a very simple embodiment of the present invention, for instance, the searching algorithm may simply develop a search string (a set of words) to plug into a pre-existing search engine. In this simple embodiment, the system will identify keywords derived from the subject matter of the media content using the tools discussed above. Then the ontology is consulted to identify related concepts and/or keywords. For instance, in the football example above, the ontology would disclose that an "interception" is a subcategory or child concept of the concept/keyword "turnover". It also would disclose that sister forms of turnover to an "interception" include "fumbles" and "safeties". The algorithm may also go down one level in the ontology to determine child concepts/keywords within the category of "interception" that may be useful search terms to add to the search string. Even further, the algorithm may look for other relations, rules, and/or restrictions, such as "interception" "may happen" on a "tipped pass," etc.
- In a slightly more complex embodiment, the various concepts/keywords developed through analysis of the ontology may be given different weights in terms of finding relevant documents. For instance, in a simple embodiment of this feature, those concepts/keywords not found directly through the subject matter analysis of the asset or segment being consumed, but identified as related concepts via the ontology may be given a lower weight in the search algorithm. In more complex embodiments, concepts from one level above may be weighed differently than sister concepts within the same level, which may be weighed differently than child concepts/keywords found by looking down one level in the ontology.
- Of course, the present invention is not limited to simply identifying related keywords for insertion into a search string. The present invention can be utilized to identify related concepts/keywords within the ontology and then incorporate those concepts/keywords in any type of searching algorithm, not just keyword searching.
- Another way to use the present invention in connection with a system such as the system disclosed in
US patent application number 12/274,452 -
Figure 3 is a flowchart illustrating process flow in accordance with the principles of the present invention in an exemplary embodiment of an automated search feature such as the one disclosed in aforementionedUS Patent Application No. 12/274,452 step 301, an ontology is created specifically adapted to a specific type of media content. The overall ontology is adapted to its particular task, e.g., a subscription-based television service provider would adapt its ontology for multimedia content, and specifically the typical television programming type of multimedia content. On the other hand, a music service provider, e.g., an online radio station, might use a very different ontology specifically adapted for audio content, and more specifically music content. Furthermore, the ontology preferably is developed with different portions thereof based on different knowledge domains, each particularly adapted to a different specific knowledge domain. Again, using the television network service provider as an example, the ontology can have within it different portions specifically adapted for sports games, cooking shows, situation comedies, reality shows, game shows, etc. If the larger domain is music, the specific knowledge domains might instead be rock, jazz, rhythm & blues and classical. - In any event, next, in
step 302, a database is built storing the subject matter information for a plurality of media items. Furthermore, the ontology is populated with the media items, i.e., the various media items are classified within the ontology. The subject matter information and classification information for the library of media items can be collected, for instance, in the ways described hereinabove, including speech recognition analysis, OCR analysis, closed caption analysis, metadata analysis, audio analytics, video analytics, external data, etc. - Next, in
step 303, the media item that is currently being consumed is analyzed using any or all of the various aforementioned technologies to determine the subject matter of that media item. Next, instep 304, the item being consumed is classified within the ontology. Instep 305, the ontology is analyzed to determine additional concepts and/or keywords relevant to the subject matter information for the media item being consumed. Next, instep 306, a search is formulated for other media items having similar subject matter information to the item being consumed using the subject matter information collected directly from the media item as well as the related concepts and key words derived from the analysis of the ontology. - Finally, in
step 307, the search is performed and the viewer is presented with the results of the search. - It should be understood that the exemplary embodiments disclosed herein in connection with automated searching for related content to a media item being consumed by a viewer is merely exemplary. The concepts of the present invention can be applied in many other contexts. For instance, it can be applied to a search, such as an Internet search, developed by a human user when there is an available ontology relevant to the knowledge domain of the search. Particularly, the search terms or keywords input into a search engine by a human user can be run through that ontology to identify additional related terms according to the ontology and then an enhanced search string can be developed for use in performing the actual search (with or without making the human user aware of the modification).
- In yet another embodiment, the ontology can be used to determine the relevance to each other of any two or more media segments already classified within the ontology. This embodiment might be useful in connection with tasks such as automatically generating playlists pertaining to a particular subject matter.
- Even further, it is not even necessary that any content be previously indexed within the ontology. An ontology itself (completely devoid of anything actually being indexed thereunder) would still provide robust information as to the relatedness of concepts within the ontology. Therefore, an ontology can be used to help improve the parameters used to search for content relevant to any particular topic within the knowledge domain of the ontology in the complete absence of any content actually being indexed under the ontology. The existence of content indexed actually under the ontology would make it easier to locate and identify relevant content since the act of identifying other related concepts within the ontology would inherently also identify the corresponding content indexed under those related concepts in the ontology.
- Furthermore, it should be understood by those of skill in the art that, while most of the embodiments discussed hereinabove use exemplary media units of segments at the sub-asset level, this is merely exemplary. The present invention can be used in connection with the classification, searching, and identifying of media items in units of any size (both physically and conceptually), including the sub-asset level, the asset level, or any other units.
- In at least one preferred embodiment of the invention, all of the media items 101 are stored in a digital memory as digital files. The ontology and the segment database also are stored in a computer or other digital memory. The various subject matter information gathering modules are preferably implemented as software routines running on a general or special purpose digital processing device. However, the processes also could be implemented in any of a number of other reasonable manners, including, but not limited to, integrated circuits, combinational logic circuits, field programmable gate arrays, analog circuitry, microprocessors, state machines, and/or combinations of any of the above. The mechanisms for formulating the search strategy as well as the mechanism for performing the search also are preferably implemented as software routines, but also could be implemented in any of the aforementioned manners.
- By designing the ontology, the subject matter information gathering process and/or the attribute/value pairs for the segment database particularly for a plurality of different specific knowledge domains based on a robust knowledge of each such knowledge domain (e.g., cooking, football, sitcoms, law dramas), one can provide much richer and more robust search and retrieval functionality for users.
- The ontology 105 can be continuously refined as types of programming, products, services, demographics, etc. are developed or themselves become more refined.
- Having thus described a few particular embodiments of the invention, various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements as are made obvious by this disclosure are intended to be part of this description though not expressly stated herein, and are intended to be within the scope of the invention. Accordingly, the foregoing description is by way of example only, and not limiting. The invention is limited only as defined in the following claims and equivalents thereto.
- There follows a selection of numbered features relating to one or more embodiments of the invention. Where a numbered feature refers to an earlier numbered feature, then all of the elements of that earlier numbered feature are to be considered included within the referring feature. Applicant gives notice that it reserves the right to claim any of the below features either singly or in any combination.
- Feature 1. A method of identifying media items relevant to a selected subject matter, each media item comprising at least one of audio content and video content, the method comprising the steps of:
- selecting a subject matter;
- determining a classification within an ontology of the selected subject matter;
- analyzing the ontology to identify at least one related subject matter to the selected subject matter according to the ontology; and
- performing a search for media items relevant to the selected subject matter as a function of at least the related subject matter.
- Feature 2. A method of identifying media items relevant to a first media item as a function of subject matter, the method comprising the steps of:
- determining a subject matter of a first media item, the first media item comprising at least one of audio content and video content;
- classifying the first media item within an ontology as a function of its subject matter;
- analyzing the ontology to identify other, related subject matter to the subject matter of the selected media item; and
- performing a search for other media items relevant to the subject matter of the first media item as a function of at least the other, related subject matter according to the ontology.
- Feature 3. The method of any of the preceding features further comprising the step of:
- indexing a plurality of media items comprising at least one of audio content and video content within the ontology as a function of their subject matter.
- Feature 4. The method of any of the preceding features wherein the step of performing the search further comprises searching for media items, indexed under the related subject matter within the ontology.
- Feature 5. The method of any of the preceding features wherein the plurality of media items comprises segments of media assets at the sub-asset level, each segment having a cohesive subject matter.
- Feature 6. The method of any of the preceding features wherein the step of analyzing the ontology to identify related subject matter comprises identifying a parent subject matter to the subject matter of the selected media item.
- Feature 7. The method of any of the preceding features wherein the step of analyzing the ontology to identify related subject matter comprises identifying a sister subject matter of the subject matter of the selected media item.
- Feature 8. The method of any of the preceding features wherein the ontology has different portions specifically adapted to different knowledge domains and the first media item comprises a segment of a media asset and the segment is classified within a portion of the ontology corresponding to the knowledge domain of the media asset of which the segment is a part.
- Feature 9. The method of feature 8 wherein the step of determining the subject matter comprises selecting a specific process for determining the subject matter as a function of a knowledge domain of the media asset.
- Feature 10. The method of feature 8 or 9 further comprising the step of:
- determining the knowledge domain of the segment.
- Feature 11. The method of any of the preceding features wherein the media assets are multimedia assets.
- Feature 12. The method of feature 11 wherein the media assets comprise television programming.
- Feature13. The method of any of the preceding features wherein the step of determining the subject matter of the first media item comprises at least one of analyzing a closed captioning stream within the media item, analyzing audio within the media item for non-speech subject matter information, performing speech recognition on the audio within the media item, performing optical character recognition on the video of the media item, performing video analytics on video of the media item, and analyzing metadata within the media item for subject matter information.
- Feature 14. The method of any of the preceding features wherein the first media item is a segment of a media asset, the segment having a cohesive subject.
- Feature 15. A computer program product for identifying media items relevant to a first media item as a function of their subject matter comprising :
- (1) computer executable instructions for determining a subject matter of a first media item, the first media item comprising at least one of audio content and video content;
- (2) computer executable instructions for classifying the first media item within the ontology as a function of its subject matter;
- (3) computer executable instructions for analyzing the ontology to identify other, related subject matter to the subject matter of the first media item;
- (4) computer executable instructions for formulating a search strategy for media items relevant to the subject matter of the first media item, the search strategy being a function of at least the subject matter related to the subject matter of the first media item according to the ontology; and
- (5) computer executable instructions for performing a search for media items relevant to the first media item using the search strategy.
- Feature 16. The computer program product of feature 15 wherein the ontology has different portions specifically adapted to different knowledge domains and the selected media item comprises a segment of a media asset and the segment is classified within a portion of the ontology corresponding to the knowledge domain of the media asset of which the segment is a part.
- Feature 17. The computer program product of feature 16 wherein the step of determining subject matter comprises selecting a specific process for determining subject matter as a function of a knowledge domain of the media asset.
- Feature 18. The computer program product of feature 16 further comprising the step of:
- determining the knowledge domain of the media asset.
- Feature 19. The computer program product of any of features 15 to 18 further comprising the steps of:
- indexing a plurality of media items comprising at least one of audio content and video content within an ontology as a function of their subject matter; and
- wherein the step of performing the search further comprises searching for media items classified under the related subject matter within the ontology.
- Feature 20. The computer program product of any of features 15 to 19 wherein the media items comprise television programming.
- Feature 21. The computer program product of any of features 15 to 20 wherein the step of analyzing the ontology to identify related subject matter comprises identifying at least one of a parent subject matter and a sister subject matter to the subject matter of the selected media item.
Claims (15)
- A method of identifying media items relevant to a selected subject matter, each media item comprising at least one of audio content and video content, the method comprising the steps of:selecting a subject matter;determining a classification within an ontology of the selected subject matter;analyzing the ontology to identify at least one related subject matter to the selected subject matter according to the ontology; andperforming a search for media items relevant to the selected subject matter as a function of at least the related subject matter.
- A method of identifying media items relevant to a first media item as a function of subject matter, the method comprising the steps of:determining a subject matter of a first media item, the first media item comprising at least one of audio content and video content;classifying the first media item within an ontology as a function of its subject matter;analyzing the ontology to identify other, related subject matter to the subject matter of the selected media item; andperforming a search for other media items relevant to the subject matter of the first media item as a function of at least the other, related subject matter according to the ontology.
- The method of any of the preceding claims, further comprising the step of:indexing a plurality of media items comprising at least one of audio content and video content within the ontology as a function of their subject matter.
- The method of any of the preceding claims wherein the step of performing the search further comprises searching for media items, indexed under the related subject matter within the ontology.
- The method of any of the preceding claims wherein the plurality of media items comprises segments of media assets at the sub-asset level, each segment having a cohesive subject matter.
- The method of any of the preceding claims wherein the step of analyzing the ontology to identify related subject matter comprises identifying a parent subject matter and/or a sister subject matter to the subject matter of the selected media item.
- The method of any of the preceding claims wherein the ontology has different portions specifically adapted to different knowledge domains and the first media item comprises a segment of a media asset and the segment is classified within a portion of the ontology corresponding to the knowledge domain of the media asset of which the segment is a part.
- The method of claim 7 wherein the step of determining the subject matter comprises selecting a specific process for determining the subject matter as a function of a knowledge domain of the media asset.
- The method of claims 7 or 8 further comprising the step of:determining the knowledge domain of the segment.
- The method of any of the preceding claims wherein the media assets are multimedia assets, and preferably comprise television programming.
- The method of any of the preceding claims wherein the step of determining the subject matter of the first media item comprises at least one of analyzing a closed captioning stream within the media item, analyzing audio within the media item for non-speech subject matter information, performing speech recognition on the audio within the media item, performing optical character recognition on the video of the media item, performing video analytics on video of the media item, and analyzing metadata within the media item for subject matter information.
- The method of any of the preceding claims wherein the first media item is a segment of a media asset, the segment having a cohesive subject.
- A computer program or suite of computer programs, so arranged such that when executed by a computer it/they cause the computer to perform the method of any of the preceding claims.
- A computer program product storing a computer program or at least one of the suite of computer programs according to claim 13.
- An apparatus for identifying media items relevant to a first media item as a function of subject matter, the apparatus comprising:means for determining a subject matter of a first media item, the first media item comprising at least one of audio content and video content;means for classifying the first media item within an ontology as a function of its subject matter;means for analyzing the ontology to identify other, related subject matter to the subject matter of the selected media item; andmeans for performing a search for other media items relevant to the subject matter of the first media item as a function of at least the other, related subject matter according to the ontology.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/343,790 US8713016B2 (en) | 2008-12-24 | 2008-12-24 | Method and apparatus for organizing segments of media assets and determining relevance of segments to a query |
Publications (1)
Publication Number | Publication Date |
---|---|
EP2202647A1 true EP2202647A1 (en) | 2010-06-30 |
Family
ID=41788932
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP09180762A Ceased EP2202647A1 (en) | 2008-12-24 | 2009-12-24 | Method and apparatus for organizing segments of media assets and determining relevance of segments to a query |
Country Status (3)
Country | Link |
---|---|
US (5) | US8713016B2 (en) |
EP (1) | EP2202647A1 (en) |
CA (1) | CA2689376C (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3073743A1 (en) * | 2015-03-24 | 2016-09-28 | Facebook, Inc. | Systems and methods for providing playback of selected video segments |
US9519719B2 (en) * | 2015-02-10 | 2016-12-13 | International Business Machines Corporation | Resource management in a presentation environment |
CN106372104A (en) * | 2016-08-19 | 2017-02-01 | 中国科学院计算机网络信息中心 | Spatiotemporal delay correlation visualization method |
US9824279B2 (en) | 2015-03-24 | 2017-11-21 | Facebook, Inc. | Systems and methods for providing playback of selected video segments |
US9846696B2 (en) * | 2012-02-29 | 2017-12-19 | Telefonaktiebolaget Lm Ericsson (Publ) | Apparatus and methods for indexing multimedia content |
US9886591B2 (en) | 2015-02-10 | 2018-02-06 | International Business Machines Corporation | Intelligent governance controls based on real-time contexts |
US20200013380A1 (en) * | 2018-07-09 | 2020-01-09 | Tree Goat Media, LLC | Systems and methods for transforming digital audio content into visual topic-based segments |
Families Citing this family (42)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9442933B2 (en) | 2008-12-24 | 2016-09-13 | Comcast Interactive Media, Llc | Identification of segments within audio, video, and multimedia items |
US8713016B2 (en) | 2008-12-24 | 2014-04-29 | Comcast Interactive Media, Llc | Method and apparatus for organizing segments of media assets and determining relevance of segments to a query |
US8176043B2 (en) | 2009-03-12 | 2012-05-08 | Comcast Interactive Media, Llc | Ranking search results |
US20100274667A1 (en) * | 2009-04-24 | 2010-10-28 | Nexidia Inc. | Multimedia access |
US8533223B2 (en) | 2009-05-12 | 2013-09-10 | Comcast Interactive Media, LLC. | Disambiguation and tagging of entities |
KR20100132705A (en) * | 2009-06-10 | 2010-12-20 | 삼성전자주식회사 | Method for providing contents list and multimedia apparatus applying the same |
US9892730B2 (en) | 2009-07-01 | 2018-02-13 | Comcast Interactive Media, Llc | Generating topic-specific language models |
JP2013529331A (en) * | 2010-04-30 | 2013-07-18 | トムソン ライセンシング | Automatic image discovery and recommendation for viewing television content |
JP4965746B2 (en) * | 2010-07-02 | 2012-07-04 | 隆敏 柳瀬 | Logical operation system |
US8612477B2 (en) | 2010-09-24 | 2013-12-17 | Aol Inc. | Systems and methods for customized electronic communications |
US20120101869A1 (en) * | 2010-10-25 | 2012-04-26 | Robert Manganelli | Media management system |
WO2013040355A1 (en) * | 2011-09-14 | 2013-03-21 | Greg Green | Mass customization of media and statistical projection from mass customized media |
US9069844B2 (en) * | 2011-11-02 | 2015-06-30 | Sap Se | Facilitating extraction and discovery of enterprise services |
US8924996B2 (en) * | 2011-11-08 | 2014-12-30 | Verizon Patent And Licensing Inc. | Session manager |
US10681427B2 (en) * | 2012-03-14 | 2020-06-09 | Arris Enterprises Llc | Sentiment mapping in a media content item |
US20160148650A1 (en) * | 2014-11-24 | 2016-05-26 | Vixs Systems, Inc. | Video processing system with custom chaptering and methods for use therewith |
US9552124B2 (en) * | 2012-04-27 | 2017-01-24 | Mobitv, Inc. | Character based search and discovery of media content |
US9177289B2 (en) | 2012-05-03 | 2015-11-03 | Sap Se | Enhancing enterprise service design knowledge using ontology-based clustering |
US20130304871A1 (en) * | 2012-05-11 | 2013-11-14 | Fred Silverman | Continually Updating a Channel of Aggregated and Curated Media Content Using Metadata |
US9633015B2 (en) * | 2012-07-26 | 2017-04-25 | Telefonaktiebolaget Lm Ericsson (Publ) | Apparatus and methods for user generated content indexing |
US10083618B2 (en) * | 2012-08-21 | 2018-09-25 | Jacob UKELSON | System and method for crowd sourced multi-media lecture capture, sharing and playback |
CN104871245A (en) * | 2012-12-18 | 2015-08-26 | 汤姆逊许可公司 | Method, apparatus and system for indexing content based on time information |
US9680689B2 (en) * | 2013-02-14 | 2017-06-13 | Comcast Cable Communications, Llc | Fragmenting media content |
AU2014262533A1 (en) | 2013-05-10 | 2015-11-26 | Uberfan, Llc | Event-related media management system |
WO2014185834A1 (en) | 2013-05-14 | 2014-11-20 | Telefonaktiebolaget L M Ericsson (Publ) | Search engine for textual content and non-textual content |
US9256687B2 (en) | 2013-06-28 | 2016-02-09 | International Business Machines Corporation | Augmenting search results with interactive search matrix |
WO2015030645A1 (en) | 2013-08-29 | 2015-03-05 | Telefonaktiebolaget L M Ericsson (Publ) | Methods, computer program, computer program product and indexing systems for indexing or updating index |
CN105493436B (en) | 2013-08-29 | 2019-09-10 | 瑞典爱立信有限公司 | For distributing method, the Content owner's equipment of content item to authorized user |
US9223833B2 (en) * | 2013-12-02 | 2015-12-29 | Qbase, LLC | Method for in-loop human validation of disambiguated features |
US20150312652A1 (en) * | 2014-04-24 | 2015-10-29 | Microsoft Corporation | Automatic generation of videos via a segment list |
CN104317825B (en) * | 2014-09-30 | 2017-06-20 | 武汉天量数据技术有限公司 | A kind of knowledge point quantitative analysis method and system |
US9854202B2 (en) * | 2014-12-11 | 2017-12-26 | Arris Enterprises Llc | Processing segments of closed-caption text using external sources |
US9544704B1 (en) * | 2015-07-16 | 2017-01-10 | Avaya Inc. | System and method for evaluating media segments for interestingness |
US10628455B2 (en) * | 2015-12-01 | 2020-04-21 | King.Com Ltd. | Method and apparatus for processing data |
US9654360B1 (en) | 2016-02-29 | 2017-05-16 | Wowza Media Systems, LLC | Coordinating analytics between media player and server |
US10116537B2 (en) | 2016-02-29 | 2018-10-30 | Wowza Media Systems, LLC | Media player analytics |
CN106354827B (en) * | 2016-08-31 | 2021-04-13 | Tcl科技集团股份有限公司 | Media asset data integration method and system |
US11042574B2 (en) * | 2016-10-05 | 2021-06-22 | International Business Machines Corporation | Contextual enhancing of content participation in search results |
US10277953B2 (en) * | 2016-12-06 | 2019-04-30 | The Directv Group, Inc. | Search for content data in content |
US10679626B2 (en) * | 2018-07-24 | 2020-06-09 | Pegah AARABI | Generating interactive audio-visual representations of individuals |
US11301629B2 (en) * | 2019-08-21 | 2022-04-12 | International Business Machines Corporation | Interleaved conversation concept flow enhancement |
US11640424B2 (en) * | 2020-08-18 | 2023-05-02 | Dish Network L.L.C. | Methods and systems for providing searchable media content and for searching within media content |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6038560A (en) * | 1997-05-21 | 2000-03-14 | Oracle Corporation | Concept knowledge base search and retrieval system |
US20030061028A1 (en) * | 2001-09-21 | 2003-03-27 | Knumi Inc. | Tool for automatically mapping multimedia annotations to ontologies |
US7177881B2 (en) | 2003-06-23 | 2007-02-13 | Sony Corporation | Network media channels |
Family Cites Families (221)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB244875A (en) | 1924-10-31 | 1925-12-31 | V L Oil Processes Ltd | Improvements in or relating to the cracking and separating of mineral oils |
US4227177A (en) | 1978-04-27 | 1980-10-07 | Dialog Systems, Inc. | Continuous speech recognition method |
US5761655A (en) | 1990-06-06 | 1998-06-02 | Alphatronix, Inc. | Image file storage and retrieval system |
GB9100732D0 (en) | 1991-01-14 | 1991-02-27 | Xerox Corp | A data access system |
US5623690A (en) | 1992-06-03 | 1997-04-22 | Digital Equipment Corporation | Audio/video storage and retrieval for multimedia workstations by interleaving audio and video data in data file |
US5682326A (en) | 1992-08-03 | 1997-10-28 | Radius Inc. | Desktop digital video processing system |
US5530859A (en) | 1993-05-10 | 1996-06-25 | Taligent, Inc. | System for synchronizing a midi presentation with presentations generated by other multimedia streams by means of clock objects |
ATE409907T1 (en) | 1993-06-03 | 2008-10-15 | Network Appliance Inc | METHOD AND DEVICE FOR DESCRIBING ANY AREAS OF A FILE SYSTEM |
US5560007A (en) | 1993-06-30 | 1996-09-24 | Borland International, Inc. | B-tree key-range bit map index optimization of database queries |
US5594897A (en) | 1993-09-01 | 1997-01-14 | Gwg Associates | Method for retrieving high relevance, high quality objects from an overall source |
US5553281A (en) | 1994-03-21 | 1996-09-03 | Visual F/X, Inc. | Method for computer-assisted media processing |
US5521841A (en) | 1994-03-31 | 1996-05-28 | Siemens Corporate Research, Inc. | Browsing contents of a given video sequence |
US5493677A (en) | 1994-06-08 | 1996-02-20 | Systems Research & Applications Corporation | Generation, archiving, and retrieval of digital images with evoked suggestion-set captions and natural language interface |
US5576755A (en) | 1994-10-28 | 1996-11-19 | Davis; Bruce | System and method for verification of electronic television program guide data |
JP3545824B2 (en) | 1995-02-21 | 2004-07-21 | 富士通株式会社 | Data retrieval device |
US5649182A (en) | 1995-03-17 | 1997-07-15 | Reitz; Carl A. | Apparatus and method for organizing timeline data |
US5729741A (en) | 1995-04-10 | 1998-03-17 | Golden Enterprises, Inc. | System for storage and retrieval of diverse types of information obtained from different media sources which includes video, audio, and text transcriptions |
US5799315A (en) | 1995-07-07 | 1998-08-25 | Sun Microsystems, Inc. | Method and apparatus for event-tagging data files automatically correlated with a time of occurence in a computer system |
US5844279A (en) * | 1995-09-14 | 1998-12-01 | Kabushiki Kaisha Toshiba | Single-electron semiconductor device |
US5717914A (en) | 1995-09-15 | 1998-02-10 | Infonautics Corporation | Method for categorizing documents into subjects using relevance normalization for documents retrieved from an information retrieval system in response to a query |
US5737734A (en) | 1995-09-15 | 1998-04-07 | Infonautics Corporation | Query word relevance adjustment in a search of an information retrieval system |
US5742816A (en) | 1995-09-15 | 1998-04-21 | Infonautics Corporation | Method and apparatus for identifying textual documents and multi-mediafiles corresponding to a search topic |
US5640553A (en) | 1995-09-15 | 1997-06-17 | Infonautics Corporation | Relevance normalization for documents retrieved from an information retrieval system in response to a query |
US5737495A (en) | 1995-09-29 | 1998-04-07 | Intel Corporation | Method and apparatus for managing multimedia data files in a computer network by streaming data files into separate streams based on file attributes |
US6370543B2 (en) | 1996-05-24 | 2002-04-09 | Magnifi, Inc. | Display of media previews |
US6374260B1 (en) | 1996-05-24 | 2002-04-16 | Magnifi, Inc. | Method and apparatus for uploading, indexing, analyzing, and searching media content |
US5765150A (en) | 1996-08-09 | 1998-06-09 | Digital Equipment Corporation | Method for statistically projecting the ranking of information |
US5928330A (en) | 1996-09-06 | 1999-07-27 | Motorola, Inc. | System, device, and method for streaming a multimedia file |
US5956729A (en) | 1996-09-06 | 1999-09-21 | Motorola, Inc. | Multimedia file, supporting multiple instances of media types, and method for forming same |
US20030093790A1 (en) | 2000-03-28 | 2003-05-15 | Logan James D. | Audio and video program recording, editing and playback systems using metadata |
US6058392A (en) | 1996-11-18 | 2000-05-02 | Wesley C. Sampson Revocable Trust | Method for the organizational indexing, storage, and retrieval of data according to data pattern signatures |
US6278992B1 (en) | 1997-03-19 | 2001-08-21 | John Andrew Curtis | Search engine using indexing method for storing and retrieving data |
US6167377A (en) | 1997-03-28 | 2000-12-26 | Dragon Systems, Inc. | Speech recognition language models |
BR9714608B1 (en) | 1997-04-14 | 2012-05-29 | method for forming a composite program guide for display by comparing data from multiple sources. | |
US5937422A (en) | 1997-04-15 | 1999-08-10 | The United States Of America As Represented By The National Security Agency | Automatically generating a topic description for text and searching and sorting text by topic using the same |
US5982369A (en) | 1997-04-21 | 1999-11-09 | Sony Corporation | Method for displaying on a screen of a computer system images representing search results |
US5924090A (en) | 1997-05-01 | 1999-07-13 | Northern Light Technology Llc | Method and apparatus for searching a database of records |
US5845279A (en) | 1997-06-27 | 1998-12-01 | Lucent Technologies Inc. | Scheduling resources for continuous media databases |
US6463444B1 (en) | 1997-08-14 | 2002-10-08 | Virage, Inc. | Video cataloger system with extensibility |
US6567980B1 (en) | 1997-08-14 | 2003-05-20 | Virage, Inc. | Video cataloger system with hyperlinked output |
US6360234B2 (en) | 1997-08-14 | 2002-03-19 | Virage, Inc. | Video cataloger system with synchronized encoders |
US6052657A (en) | 1997-09-09 | 2000-04-18 | Dragon Systems, Inc. | Text segmentation and identification of topic using language models |
US6055543A (en) | 1997-11-21 | 2000-04-25 | Verano | File wrapper containing cataloging information for content searching across multiple platforms |
US6363380B1 (en) | 1998-01-13 | 2002-03-26 | U.S. Philips Corporation | Multimedia computer system with story segmentation capability and operating program therefor including finite automation video parser |
US6356920B1 (en) | 1998-03-09 | 2002-03-12 | X-Aware, Inc | Dynamic, hierarchical data exchange system |
US6418431B1 (en) | 1998-03-30 | 2002-07-09 | Microsoft Corporation | Information retrieval and speech recognition based on language models |
US6698020B1 (en) | 1998-06-15 | 2004-02-24 | Webtv Networks, Inc. | Techniques for intelligent video ad insertion |
DE69907829T2 (en) * | 1998-09-03 | 2004-04-01 | Ricoh Co., Ltd. | Storage media with video or audio index information, management procedures and retrieval procedures for video or audio information and video retrieval system |
US6366296B1 (en) | 1998-09-11 | 2002-04-02 | Xerox Corporation | Media browser using multimodal analysis |
US6188976B1 (en) | 1998-10-23 | 2001-02-13 | International Business Machines Corporation | Apparatus and method for building domain-specific language models |
US6343294B1 (en) | 1998-12-15 | 2002-01-29 | International Business Machines Corporation | Data file editor for multiple data subsets |
US7134074B2 (en) | 1998-12-25 | 2006-11-07 | Matsushita Electric Industrial Co., Ltd. | Data processing method and storage medium, and program for causing computer to execute the data processing method |
US6345253B1 (en) | 1999-04-09 | 2002-02-05 | International Business Machines Corporation | Method and apparatus for retrieving audio information using primary and supplemental indexes |
US6937766B1 (en) | 1999-04-15 | 2005-08-30 | MATE—Media Access Technologies Ltd. | Method of indexing and searching images of text in video |
US6415434B1 (en) | 1999-06-18 | 2002-07-02 | Hewlett-Packard Company | Apparatus and method for a runtime method overloading resolver |
CA2377941A1 (en) | 1999-06-28 | 2001-01-04 | United Video Properties, Inc. | Interactive television program guide system and method with niche hubs |
US7181438B1 (en) | 1999-07-21 | 2007-02-20 | Alberti Anemometer, Llc | Database access system |
US6546385B1 (en) | 1999-08-13 | 2003-04-08 | International Business Machines Corporation | Method and apparatus for indexing and searching content in hardcopy documents |
GB2354104A (en) | 1999-09-08 | 2001-03-14 | Sony Uk Ltd | An editing method and system |
US6963867B2 (en) | 1999-12-08 | 2005-11-08 | A9.Com, Inc. | Search query processing to provide category-ranked presentation of search results |
US6675174B1 (en) | 2000-02-02 | 2004-01-06 | International Business Machines Corp. | System and method for measuring similarity between a set of known temporal media segments and a one or more temporal media streams |
US6882793B1 (en) | 2000-06-16 | 2005-04-19 | Yesvideo, Inc. | Video processing system |
US6580437B1 (en) | 2000-06-26 | 2003-06-17 | Siemens Corporate Research, Inc. | System for organizing videos based on closed-caption information |
US6545209B1 (en) | 2000-07-05 | 2003-04-08 | Microsoft Corporation | Music content characteristic identification and matching |
US7490092B2 (en) | 2000-07-06 | 2009-02-10 | Streamsage, Inc. | Method and system for indexing and searching timed media information based upon relevance intervals |
US7206775B2 (en) | 2000-07-06 | 2007-04-17 | Microsoft Corporation | System and methods for the automatic transmission of new, high affinity media |
EP1405212B1 (en) | 2000-07-06 | 2013-12-18 | Streamsage, Inc. | Method and system for indexing and searching timed media information based upon relevance intervals |
US6957184B2 (en) | 2000-07-20 | 2005-10-18 | Microsoft Corporation | Context free grammar engine for speech recognition system |
US7464086B2 (en) | 2000-08-01 | 2008-12-09 | Yahoo! Inc. | Metatag-based datamining |
US6789088B1 (en) * | 2000-10-19 | 2004-09-07 | Lg Electronics Inc. | Multimedia description scheme having weight information and method for displaying multimedia |
US20020087315A1 (en) | 2000-12-29 | 2002-07-04 | Lee Victor Wai Leung | Computer-implemented multi-scanning language method and system |
US6792426B2 (en) | 2001-01-10 | 2004-09-14 | International Business Machines Corporation | Generic servlet for browsing EJB entity beans |
US6678635B2 (en) | 2001-01-23 | 2004-01-13 | Intel Corporation | Method and system for detecting semantic events |
EP1241587A1 (en) | 2001-03-12 | 2002-09-18 | THOMSON multimedia | Method for processing content data |
US6847974B2 (en) | 2001-03-26 | 2005-01-25 | Us Search.Com Inc | Method and apparatus for intelligent data assimilation |
US20030163815A1 (en) | 2001-04-06 | 2003-08-28 | Lee Begeja | Method and system for personalized multimedia delivery service |
US20030101170A1 (en) | 2001-05-25 | 2003-05-29 | Joseph Edelstein | Data query and location through a central ontology model |
US6993532B1 (en) | 2001-05-30 | 2006-01-31 | Microsoft Corporation | Auto playlist generator |
US7003503B2 (en) | 2001-06-07 | 2006-02-21 | Idealswork Inc. | Ranking items |
KR100492093B1 (en) * | 2001-07-13 | 2005-06-01 | 삼성전자주식회사 | System and method for providing summary video information of video data |
JP2003036093A (en) | 2001-07-23 | 2003-02-07 | Japan Science & Technology Corp | Speech input retrieval system |
JP4342753B2 (en) | 2001-08-10 | 2009-10-14 | 株式会社リコー | Document search apparatus, document search method, program, and computer-readable storage medium |
US6901364B2 (en) | 2001-09-13 | 2005-05-31 | Matsushita Electric Industrial Co., Ltd. | Focused language models for improved speech input of structured documents |
US7099293B2 (en) | 2002-05-01 | 2006-08-29 | Stmicroelectronics, Inc. | Buffer-less de-skewing for symbol combination in a CDMA demodulator |
US6567960B2 (en) * | 2001-10-09 | 2003-05-20 | Hewlett-Packard Development L.P. | System for improving circuit simulations by utilizing a simplified circuit model based on effective capacitance and inductance values |
GB2381688B (en) | 2001-11-03 | 2004-09-22 | Dremedia Ltd | Time ordered indexing of audio-visual data |
US20030135582A1 (en) | 2001-12-21 | 2003-07-17 | Docomo Communications Laboratories Usa, Inc. | Context aware search service |
US6772147B2 (en) | 2002-02-26 | 2004-08-03 | Sony Corporation | System and method for effectively implementing a personal channel for interactive television |
US6694323B2 (en) | 2002-04-25 | 2004-02-17 | Sybase, Inc. | System and methodology for providing compact B-Tree |
US7836466B2 (en) | 2002-06-06 | 2010-11-16 | Microsoft Corporation | Methods and systems for generating electronic program guides |
WO2004010335A1 (en) | 2002-07-23 | 2004-01-29 | Samsung Electronics Co., Ltd. | Index structure of metadata, method for providing indices of metadata, and metadata searching method and apparatus using the indices of metadata |
US7584102B2 (en) | 2002-11-15 | 2009-09-01 | Scansoft, Inc. | Language model for use in speech recognition |
US20040167883A1 (en) | 2002-12-06 | 2004-08-26 | Attensity Corporation | Methods and systems for providing a service for producing structured data elements from free text sources |
US7870279B2 (en) * | 2002-12-09 | 2011-01-11 | Hrl Laboratories, Llc | Method and apparatus for scanning, personalizing, and casting multimedia data streams via a communication network and television |
AU2002953555A0 (en) | 2002-12-23 | 2003-01-16 | Canon Kabushiki Kaisha | Method for presenting hierarchical data |
GB2399427A (en) | 2003-03-12 | 2004-09-15 | Canon Kk | Apparatus for and method of summarising text |
EP1462950B1 (en) | 2003-03-27 | 2007-08-29 | Sony Deutschland GmbH | Method for language modelling |
US7647344B2 (en) | 2003-05-29 | 2010-01-12 | Experian Marketing Solutions, Inc. | System, method and software for providing persistent entity identification and linking entity information in an integrated data repository |
US7162473B2 (en) | 2003-06-26 | 2007-01-09 | Microsoft Corporation | Method and system for usage analyzer that determines user accessed sources, indexes data subsets, and associated metadata, processing implicit queries based on potential interest to users |
US20050044105A1 (en) | 2003-08-19 | 2005-02-24 | Kelly Terrell | System and method for delivery of content-specific video clips |
US7499912B2 (en) | 2003-10-23 | 2009-03-03 | Hywire Ltd. | Search method using coded keys |
US20050114130A1 (en) | 2003-11-20 | 2005-05-26 | Nec Laboratories America, Inc. | Systems and methods for improving feature ranking using phrasal compensation and acronym detection |
EP2506252B1 (en) | 2003-11-21 | 2019-06-05 | Nuance Communications, Inc. | Topic specific models for text formatting and speech recognition |
US20050124320A1 (en) | 2003-12-09 | 2005-06-09 | Johannes Ernst | System and method for the light-weight management of identity and related information |
US7461089B2 (en) | 2004-01-08 | 2008-12-02 | International Business Machines Corporation | Method and system for creating profiling indices |
EP2317449A3 (en) | 2004-01-16 | 2011-07-13 | Hillcrest Laboratories, Inc. | Metadata brokering server and methods |
US7984089B2 (en) | 2004-02-13 | 2011-07-19 | Microsoft Corporation | User-defined indexing of multimedia content |
US7376642B2 (en) | 2004-03-30 | 2008-05-20 | Microsoft Corporation | Integrated full text search system and method |
US7231405B2 (en) | 2004-05-08 | 2007-06-12 | Doug Norman, Interchange Corp. | Method and apparatus of indexing web pages of a web site for geographical searchine based on user location |
TWI273449B (en) | 2004-06-18 | 2007-02-11 | Yi-Jie Wu | Computer data classification management system and method |
US7818282B2 (en) | 2004-07-02 | 2010-10-19 | International Business Machines Corporation | System and method for the support of multilingual applications |
US8078607B2 (en) | 2006-03-30 | 2011-12-13 | Google Inc. | Generating website profiles based on queries from webistes and user activities on the search results |
US20060037046A1 (en) | 2004-08-13 | 2006-02-16 | Microsoft Corporation | Aligning video data to create a comprehensive program guide |
US7925506B2 (en) | 2004-10-05 | 2011-04-12 | Inago Corporation | Speech recognition accuracy via concept to keyword mapping |
US7949675B2 (en) | 2004-10-15 | 2011-05-24 | Oracle International Corporation | Data retrieval method |
US7526506B2 (en) | 2004-10-21 | 2009-04-28 | Microsoft Corporation | Interlinking sports and television program listing metadata |
US8335753B2 (en) * | 2004-11-03 | 2012-12-18 | Microsoft Corporation | Domain knowledge-assisted information processing |
US20060167859A1 (en) | 2004-11-09 | 2006-07-27 | Verbeck Sibley Timothy J | System and method for personalized searching of television content using a reduced keypad |
US20060100898A1 (en) | 2004-11-10 | 2006-05-11 | Pearce Jonathan P | Method for booking and scheduling actors for projects using a computer network |
US7587396B2 (en) | 2004-11-24 | 2009-09-08 | Oracle International Corporation | Encoding data to be sorted |
US20060156399A1 (en) | 2004-12-30 | 2006-07-13 | Parmar Pankaj N | System and method for implementing network security using a sequestered partition |
US7680791B2 (en) | 2005-01-18 | 2010-03-16 | Oracle International Corporation | Method for sorting data using common prefix bytes |
EP1846815A2 (en) | 2005-01-31 | 2007-10-24 | Textdigger, Inc. | Method and system for semantic search and retrieval of electronic documents |
US20060256739A1 (en) | 2005-02-19 | 2006-11-16 | Kenneth Seier | Flexible multi-media data management |
WO2006099621A2 (en) | 2005-03-17 | 2006-09-21 | University Of Southern California | Topic specific language models built from large numbers of documents |
US8812473B1 (en) | 2005-06-16 | 2014-08-19 | Gere Dev. Applications, LLC | Analysis and reporting of collected search activity data over multiple search engines |
US20070011133A1 (en) | 2005-06-22 | 2007-01-11 | Sbc Knowledge Ventures, L.P. | Voice search engine generating sub-topics based on recognitiion confidence |
US8566789B2 (en) | 2005-08-25 | 2013-10-22 | Infosys Limited | Semantic-based query techniques for source code |
US8250051B2 (en) | 2005-08-26 | 2012-08-21 | Harris Corporation | System, program product, and methods to enhance media content management |
US7788266B2 (en) | 2005-08-26 | 2010-08-31 | Veveo, Inc. | Method and system for processing ambiguous, multi-term search queries |
US7672833B2 (en) | 2005-09-22 | 2010-03-02 | Fair Isaac Corporation | Method and apparatus for automatic entity disambiguation |
US20070078708A1 (en) | 2005-09-30 | 2007-04-05 | Hua Yu | Using speech recognition to determine advertisements relevant to audio content and/or audio content relevant to advertisements |
US7590536B2 (en) | 2005-10-07 | 2009-09-15 | Nuance Communications, Inc. | Voice language model adjustment based on user affinity |
US8265933B2 (en) | 2005-12-22 | 2012-09-11 | Nuance Communications, Inc. | Speech recognition system for providing voice recognition services using a conversational language model |
US20070208567A1 (en) | 2006-03-01 | 2007-09-06 | At&T Corp. | Error Correction In Automatic Speech Recognition Transcripts |
US8195650B2 (en) | 2007-02-28 | 2012-06-05 | Samsung Electronics Co., Ltd. | Method and system for providing information using a supplementary device |
US20080235209A1 (en) | 2007-03-20 | 2008-09-25 | Samsung Electronics Co., Ltd. | Method and apparatus for search result snippet analysis for query expansion and result filtering |
US8115869B2 (en) | 2007-02-28 | 2012-02-14 | Samsung Electronics Co., Ltd. | Method and system for extracting relevant information from content metadata |
US20070214123A1 (en) | 2006-03-07 | 2007-09-13 | Samsung Electronics Co., Ltd. | Method and system for providing a user interface application and presenting information thereon |
US8510453B2 (en) | 2007-03-21 | 2013-08-13 | Samsung Electronics Co., Ltd. | Framework for correlating content on a local network with information on an external network |
US8732154B2 (en) | 2007-02-28 | 2014-05-20 | Samsung Electronics Co., Ltd. | Method and system for providing sponsored information on electronic devices |
US8209724B2 (en) | 2007-04-25 | 2012-06-26 | Samsung Electronics Co., Ltd. | Method and system for providing access to information of potential interest to a user |
US9100723B2 (en) | 2006-03-07 | 2015-08-04 | Samsung Electronics Co., Ltd. | Method and system for managing information on a video recording |
US20090025054A1 (en) | 2007-07-17 | 2009-01-22 | Samsung Electronics Co., Ltd. | Method and system for access to content in a content space |
US20080221989A1 (en) | 2007-03-09 | 2008-09-11 | Samsung Electronics Co., Ltd. | Method and system for providing sponsored content on an electronic device |
US20080250010A1 (en) | 2007-04-05 | 2008-10-09 | Samsung Electronics Co., Ltd. | Method and system for determining and pre-processing potential user queries related to content in a network |
US8843467B2 (en) | 2007-05-15 | 2014-09-23 | Samsung Electronics Co., Ltd. | Method and system for providing relevant information to a user of a device in a local network |
US8200688B2 (en) | 2006-03-07 | 2012-06-12 | Samsung Electronics Co., Ltd. | Method and system for facilitating information searching on electronic devices |
US8863221B2 (en) | 2006-03-07 | 2014-10-14 | Samsung Electronics Co., Ltd. | Method and system for integrating content and services among multiple networks |
CN101034390A (en) | 2006-03-10 | 2007-09-12 | 日电(中国)有限公司 | Apparatus and method for verbal model switching and self-adapting |
EP2011017A4 (en) | 2006-03-30 | 2010-07-07 | Stanford Res Inst Int | Method and apparatus for annotating media streams |
US9135238B2 (en) | 2006-03-31 | 2015-09-15 | Google Inc. | Disambiguation of named entities |
US7792812B1 (en) | 2006-03-31 | 2010-09-07 | Netlogic Microsystems, Inc. | Search engine devices that support high speed parallel decoding of digital search tries |
US7596549B1 (en) * | 2006-04-03 | 2009-09-29 | Qurio Holdings, Inc. | Methods, systems, and products for analyzing annotations for related content |
US8019754B2 (en) | 2006-04-03 | 2011-09-13 | Needlebot Incorporated | Method of searching text to find relevant content |
US7756708B2 (en) | 2006-04-03 | 2010-07-13 | Google Inc. | Automatic language model update |
US7716361B2 (en) | 2006-05-04 | 2010-05-11 | Samsung Electronics Co., Ltd. | Method and system for providing persistent media across web pages using only browser plug-ins |
US7526486B2 (en) | 2006-05-22 | 2009-04-28 | Initiate Systems, Inc. | Method and system for indexing information about entities with respect to hierarchies |
US7921116B2 (en) * | 2006-06-16 | 2011-04-05 | Microsoft Corporation | Highly meaningful multimedia metadata creation and associations |
JP5218052B2 (en) | 2006-06-26 | 2013-06-26 | 日本電気株式会社 | Language model generation system, language model generation method, and language model generation program |
US20090271195A1 (en) | 2006-07-07 | 2009-10-29 | Nec Corporation | Speech recognition apparatus, speech recognition method, and speech recognition program |
JP4257615B2 (en) | 2006-07-14 | 2009-04-22 | ソニー株式会社 | Image processing apparatus and method, and program |
US7962937B2 (en) | 2006-08-01 | 2011-06-14 | Microsoft Corporation | Media content catalog service |
US7921106B2 (en) | 2006-08-03 | 2011-04-05 | Microsoft Corporation | Group-by attribute value in search results |
US7917492B2 (en) * | 2007-09-21 | 2011-03-29 | Limelight Networks, Inc. | Method and subsystem for information acquisition and aggregation to facilitate ontology and language-model generation within a content-search-service system |
KR100813170B1 (en) | 2006-09-27 | 2008-03-17 | 삼성전자주식회사 | Method and system for semantic event indexing by analyzing user annotation of digital photos |
WO2008053132A1 (en) | 2006-11-03 | 2008-05-08 | Nds Limited | Program guide search |
US7272558B1 (en) | 2006-12-01 | 2007-09-18 | Coveo Solutions Inc. | Speech recognition training method for audio and video file indexing on a search engine |
US8935269B2 (en) | 2006-12-04 | 2015-01-13 | Samsung Electronics Co., Ltd. | Method and apparatus for contextual search and query refinement on consumer electronics devices |
CN100449547C (en) | 2006-12-06 | 2009-01-07 | 华为技术有限公司 | Medium contents management system and method |
US7739247B2 (en) | 2006-12-28 | 2010-06-15 | Ebay Inc. | Multi-pass data organization and automatic naming |
US8898701B2 (en) | 2006-12-29 | 2014-11-25 | Verizon Patent And Licensing Inc. | Method and system for providing attribute browsing of video assets |
US20080168045A1 (en) | 2007-01-10 | 2008-07-10 | Microsoft Corporation | Content rank |
US20080183681A1 (en) | 2007-01-29 | 2008-07-31 | Samsung Electronics Co., Ltd. | Method and system for facilitating information searching on electronic devices |
US7818176B2 (en) | 2007-02-06 | 2010-10-19 | Voicebox Technologies, Inc. | System and method for selecting and presenting advertisements based on natural language processing of voice-based input |
US8112402B2 (en) | 2007-02-26 | 2012-02-07 | Microsoft Corporation | Automatic disambiguation based on a reference resource |
US7958104B2 (en) | 2007-03-08 | 2011-06-07 | O'donnell Shawn C | Context based data searching |
US7801888B2 (en) | 2007-03-09 | 2010-09-21 | Microsoft Corporation | Media content search results ranked by popularity |
US20080222105A1 (en) | 2007-03-09 | 2008-09-11 | Joseph Matheny | Entity recommendation system using restricted information tagged to selected entities |
EP1975813A1 (en) | 2007-03-31 | 2008-10-01 | Sony Deutschland Gmbh | Method for content recommendation |
US8055762B2 (en) | 2007-04-16 | 2011-11-08 | Samsung Electronics Co. Ltd | Method and system for location identification |
GB2448874A (en) | 2007-04-30 | 2008-11-05 | Hewlett Packard Development Co | Context based media recommender |
GB2448875B (en) | 2007-04-30 | 2011-06-01 | Isotek Electronics Ltd | A temperature compensated tuneable TEM mode resonator |
US8069188B2 (en) | 2007-05-07 | 2011-11-29 | Applied Technical Systems, Inc. | Database system storing a data structure that includes data nodes connected by context nodes and related method |
US9002869B2 (en) | 2007-06-22 | 2015-04-07 | Google Inc. | Machine translation for query expansion |
US20090006391A1 (en) | 2007-06-27 | 2009-01-01 | T Reghu Ram | Automatic categorization of document through tagging |
US20090006315A1 (en) | 2007-06-29 | 2009-01-01 | Sougata Mukherjea | Structured method for schema matching using multiple levels of ontologies |
US8468083B1 (en) | 2007-08-10 | 2013-06-18 | Google Inc. | Advertisement topic diversification and ranking |
US7983902B2 (en) | 2007-08-23 | 2011-07-19 | Google Inc. | Domain dictionary creation by detection of new topic words using divergence value comparison |
US20090094113A1 (en) | 2007-09-07 | 2009-04-09 | Digitalsmiths Corporation | Systems and Methods For Using Video Metadata to Associate Advertisements Therewith |
US8108411B2 (en) | 2007-09-14 | 2012-01-31 | Yahoo! Inc. | Methods and systems for merging data sets |
US8301633B2 (en) | 2007-10-01 | 2012-10-30 | Palo Alto Research Center Incorporated | System and method for semantic search |
US8909655B1 (en) | 2007-10-11 | 2014-12-09 | Google Inc. | Time based ranking |
US8594996B2 (en) | 2007-10-17 | 2013-11-26 | Evri Inc. | NLP-based entity recognition and disambiguation |
US8572087B1 (en) | 2007-10-17 | 2013-10-29 | Google Inc. | Content identification |
US8626618B2 (en) | 2007-11-14 | 2014-01-07 | Panjiva, Inc. | Using non-public shipper records to facilitate rating an entity based on public records of supply transactions |
US8301197B2 (en) | 2007-11-18 | 2012-10-30 | Qualcomm Incorporated | Method and apparatus for synchronizing contacts stored on smart card with contacts stored in an internal memory |
US7966304B2 (en) | 2007-11-30 | 2011-06-21 | Yahoo! Inc. | Enabling searching on abbreviated search terms via messaging |
US8065325B2 (en) | 2007-12-12 | 2011-11-22 | Packet Video Corp. | System and method for creating metadata |
US20090172544A1 (en) | 2007-12-31 | 2009-07-02 | United Video Properties, Inc. | Interactive media guidance application with access to multiple data sets |
US7814267B1 (en) | 2008-02-04 | 2010-10-12 | Netlogic Microsystems, Inc. | Processor with compare operations based on any of multiple compare data segments |
US20090205018A1 (en) | 2008-02-07 | 2009-08-13 | Ferraiolo David F | Method and system for the specification and enforcement of arbitrary attribute-based access control policies |
US8244721B2 (en) | 2008-02-13 | 2012-08-14 | Microsoft Corporation | Using related users data to enhance web search |
US8312022B2 (en) | 2008-03-21 | 2012-11-13 | Ramp Holdings, Inc. | Search engine optimization |
US8266307B2 (en) | 2008-05-12 | 2012-09-11 | Nokia Corporation | Method, system, and apparatus for access of network services using subscriber identities |
US20090326947A1 (en) | 2008-06-27 | 2009-12-31 | James Arnold | System and method for spoken topic or criterion recognition in digital media and contextual advertising |
US9374553B2 (en) | 2008-08-06 | 2016-06-21 | Rovi Guides, Inc. | Systems and methods for identifying and merging recorded segments belonging to the same program |
US7882143B2 (en) | 2008-08-15 | 2011-02-01 | Athena Ann Smyros | Systems and methods for indexing information for a search engine |
US20100063886A1 (en) | 2008-09-05 | 2010-03-11 | Dennis Stratton | Methods and systems for producing an entertainment product |
JP2010067175A (en) | 2008-09-12 | 2010-03-25 | Toshiba Corp | Hybrid content recommendation server, recommendation system, and recommendation method |
KR101040119B1 (en) | 2008-10-14 | 2011-06-09 | 한국전자통신연구원 | Apparatus and Method for Search of Contents |
US11832024B2 (en) | 2008-11-20 | 2023-11-28 | Comcast Cable Communications, Llc | Method and apparatus for delivering video and video-related content at sub-asset level |
US8392706B2 (en) | 2008-11-26 | 2013-03-05 | Perlustro, L.P. | Method and system for searching for, and collecting, electronically-stored information |
US8713016B2 (en) | 2008-12-24 | 2014-04-29 | Comcast Interactive Media, Llc | Method and apparatus for organizing segments of media assets and determining relevance of segments to a query |
US9442933B2 (en) | 2008-12-24 | 2016-09-13 | Comcast Interactive Media, Llc | Identification of segments within audio, video, and multimedia items |
WO2010100853A1 (en) | 2009-03-04 | 2010-09-10 | 日本電気株式会社 | Language model adaptation device, speech recognition device, language model adaptation method, and computer-readable recording medium |
US8176043B2 (en) | 2009-03-12 | 2012-05-08 | Comcast Interactive Media, Llc | Ranking search results |
US8229883B2 (en) | 2009-03-30 | 2012-07-24 | Sap Ag | Graph based re-composition of document fragments for name entity recognition under exploitation of enterprise databases |
US8533223B2 (en) | 2009-05-12 | 2013-09-10 | Comcast Interactive Media, LLC. | Disambiguation and tagging of entities |
US9892730B2 (en) | 2009-07-01 | 2018-02-13 | Comcast Interactive Media, Llc | Generating topic-specific language models |
US8996563B2 (en) | 2010-04-06 | 2015-03-31 | Tokutek, Inc. | High-performance streaming dictionary |
-
2008
- 2008-12-24 US US12/343,790 patent/US8713016B2/en active Active
-
2009
- 2009-12-24 CA CA2689376A patent/CA2689376C/en active Active
- 2009-12-24 EP EP09180762A patent/EP2202647A1/en not_active Ceased
-
2014
- 2014-03-06 US US14/199,063 patent/US9477712B2/en active Active
-
2016
- 2016-09-08 US US15/259,636 patent/US10635709B2/en active Active
-
2020
- 2020-03-09 US US16/813,051 patent/US11468109B2/en active Active
-
2022
- 2022-09-06 US US17/903,560 patent/US20230070056A1/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6038560A (en) * | 1997-05-21 | 2000-03-14 | Oracle Corporation | Concept knowledge base search and retrieval system |
US20030061028A1 (en) * | 2001-09-21 | 2003-03-27 | Knumi Inc. | Tool for automatically mapping multimedia annotations to ontologies |
US7177881B2 (en) | 2003-06-23 | 2007-02-13 | Sony Corporation | Network media channels |
Non-Patent Citations (2)
Title |
---|
LI, Y. ET AL.: "Reliable Video Clock Recognition", PATTERN RECOGNITION, 2006 |
SALTON ET AL: "Computer Evaluation of Indexing and Text Processing", JOURNAL OF THE ASSOCIATION FOR COMPUTING MACHINERY, vol. 15, no. 1, January 1968 (1968-01-01), New York, NY, USA, pages 8 - 36, XP002574323, Retrieved from the Internet <URL:http://portal.acm.org/citation.cfm?id=321439.321441> [retrieved on 20100317] * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9846696B2 (en) * | 2012-02-29 | 2017-12-19 | Telefonaktiebolaget Lm Ericsson (Publ) | Apparatus and methods for indexing multimedia content |
US10043024B2 (en) | 2015-02-10 | 2018-08-07 | International Business Machines Corporation | Intelligent governance controls based on real-time contexts |
US9519719B2 (en) * | 2015-02-10 | 2016-12-13 | International Business Machines Corporation | Resource management in a presentation environment |
US9888006B2 (en) | 2015-02-10 | 2018-02-06 | International Business Machines Corporation | Resource management in a presentation environment |
US9886591B2 (en) | 2015-02-10 | 2018-02-06 | International Business Machines Corporation | Intelligent governance controls based on real-time contexts |
US9923898B2 (en) | 2015-02-10 | 2018-03-20 | International Business Machines Corporation | Resource management in a presentation environment |
US9525693B2 (en) * | 2015-02-10 | 2016-12-20 | International Business Machines Corporation | Resource management in a presentation environment |
US10248867B2 (en) | 2015-03-24 | 2019-04-02 | Facebook, Inc. | Systems and methods for providing playback of selected video segments |
JP2018514127A (en) * | 2015-03-24 | 2018-05-31 | フェイスブック,インク. | System and method for providing playback of selected video segments |
US9824279B2 (en) | 2015-03-24 | 2017-11-21 | Facebook, Inc. | Systems and methods for providing playback of selected video segments |
EP3073743A1 (en) * | 2015-03-24 | 2016-09-28 | Facebook, Inc. | Systems and methods for providing playback of selected video segments |
US10860862B2 (en) | 2015-03-24 | 2020-12-08 | Facebook, Inc. | Systems and methods for providing playback of selected video segments |
CN106372104B (en) * | 2016-08-19 | 2020-04-21 | 中科智慧水运科技创新(广州)有限公司 | Time-space delay correlation visualization method |
CN106372104A (en) * | 2016-08-19 | 2017-02-01 | 中国科学院计算机网络信息中心 | Spatiotemporal delay correlation visualization method |
WO2020014223A1 (en) * | 2018-07-09 | 2020-01-16 | Tree Goat Media, LLC | Systems and methods for transforming digital audio content into visual topic-based segments |
US20200013380A1 (en) * | 2018-07-09 | 2020-01-09 | Tree Goat Media, LLC | Systems and methods for transforming digital audio content into visual topic-based segments |
US20240177695A1 (en) * | 2018-07-09 | 2024-05-30 | Tree Goat Media, Inc. | Systems and methods for transforming digital audio content into visual topic-based segments |
US10971121B2 (en) * | 2018-07-09 | 2021-04-06 | Tree Goat Media, Inc. | Systems and methods for transforming digital audio content into visual topic-based segments |
US20210166666A1 (en) * | 2018-07-09 | 2021-06-03 | Tree Goat Media, LLC | Systems and methods for transforming digitial audio content into visual topic-based segments |
US11749241B2 (en) | 2018-07-09 | 2023-09-05 | Tree Goat Media, Inc. | Systems and methods for transforming digitial audio content into visual topic-based segments |
Also Published As
Publication number | Publication date |
---|---|
US20140258284A1 (en) | 2014-09-11 |
US20230070056A1 (en) | 2023-03-09 |
CA2689376A1 (en) | 2010-06-24 |
US20170161284A1 (en) | 2017-06-08 |
US10635709B2 (en) | 2020-04-28 |
US11468109B2 (en) | 2022-10-11 |
US8713016B2 (en) | 2014-04-29 |
CA2689376C (en) | 2018-09-25 |
US9477712B2 (en) | 2016-10-25 |
US20200233895A1 (en) | 2020-07-23 |
US20100161580A1 (en) | 2010-06-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11468109B2 (en) | Searching for segments based on an ontology | |
CA2688921C (en) | Identification of segments within audio, video, and multimedia items | |
US10372759B2 (en) | Profile based content retrieval for recommender systems | |
EP2204748A1 (en) | Method and apparatus for advertising at the sub-asset level | |
US9253511B2 (en) | Systems and methods for performing multi-modal video datastream segmentation | |
Gao et al. | Vlogging: A survey of videoblogging technology on the web | |
KR101816113B1 (en) | Estimating and displaying social interest in time-based media | |
US20130166587A1 (en) | User Interface for Viewing Targeted Segments of Multimedia Content Based on Time-Based Metadata Search Criteria | |
US20020144293A1 (en) | Automatic video retriever genie | |
KR20100116412A (en) | Apparatus and method for providing advertisement information based on video scene | |
US20090132510A1 (en) | Device for enabling to represent content items through meta summary data, and method thereof | |
EP2524324A1 (en) | Method for navigating identifiers placed in areas and receiver implementing the method | |
Sumiyoshi et al. | CurioView: TV recommendations related to content being viewed | |
Dao et al. | A new spatio-temporal method for event detection and personalized retrieval of sports video | |
Browne et al. | Dublin City University video track experiments for TREC 2003 | |
Hölbling et al. | Content-based tag generation to enable a tag-based collaborative tv-recommendation system. | |
Takahashi et al. | Automatic video summarization of sports videos using metadata | |
Wang et al. | Exploiting content relevance and social relevance for personalized ad recommendation on internet TV | |
Ibrahim | TV Stream table of content: a new level in the hierarchical video representation | |
Babaguchi et al. | Effect of personalization on retrieval and summarization of sports video | |
Liang et al. | Personalized sports video customization using content and context analysis | |
Barbieri | Automatic summarization of narrative video | |
Rautiainen et al. | Kuukkeli-TV: Online content-based services and applications for broadcast TV with long-term user experiments | |
Albayrak et al. | Towards “Semantic IPTV” | |
Riley | The revolution will be televised: Identifying, organizing, and presenting correlations between social media and broadcast television |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO SE SI SK SM TR |
|
AX | Request for extension of the european patent |
Extension state: AL BA RS |
|
17P | Request for examination filed |
Effective date: 20101209 |
|
17Q | First examination report despatched |
Effective date: 20150119 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE APPLICATION HAS BEEN REFUSED |
|
18R | Application refused |
Effective date: 20171016 |